Fabio
c36510c21e
Some checks failed
Linux Build / Linux cpu (push) Has been cancelled
Linux Build / Linux cpu-clang (push) Has been cancelled
Linux Build / Linux vulkan (push) Has been cancelled
mac-build / macOS CPU (push) Has been cancelled
Nix Build / Nix (cpu) (push) Has been cancelled
Nix Build / Nix (vulkan) (push) Has been cancelled
windows-build / Windows CPU (push) Has been cancelled
817 lines
89 KiB
JSON
817 lines
89 KiB
JSON
{
|
||
"ming3_tts": [
|
||
{"name":"template_name","type":"text","label":"Prompt template","placeholder":"Automatic"},
|
||
{"name":"voice_id","type":"text","label":"IP voice name","placeholder":"US_James (IP template only)"},
|
||
{"name":"temperature","type":"number","label":"Temperature","default":0,"minimum":0,"step":0.1},
|
||
{"name":"guidance_scale","type":"number","label":"Guidance scale","default":2,"step":0.1},
|
||
{"name":"sigma","type":"number","label":"SDE noise scale","default":0.25,"minimum":0,"step":0.05},
|
||
{"name":"trim_silence","type":"bool","label":"Trim trailing silence","default":true}
|
||
],
|
||
"ming3_audio": [
|
||
{"name":"voice_id","type":"text","label":"Voice name","placeholder":"US_James"},
|
||
{"name":"prompt","type":"text","label":"Audio task prompt","placeholder":"Use task default"},
|
||
{"name":"system_prompt","type":"text","label":"System prompt"},
|
||
{"name":"enable_thinking","type":"bool","label":"Enable thinking","default":false},
|
||
{"name":"max_audio_tokens","type":"number","label":"Maximum acoustic patches per chunk","default":1000,"minimum":1,"step":1},
|
||
{"name":"temperature","type":"number","label":"Temperature","default":0,"minimum":0,"step":0.1},
|
||
{"name":"guidance_scale","type":"number","label":"Guidance scale","default":2,"step":0.1},
|
||
{"name":"disk_experts","scope":"session","session_option":"ming3_audio.disk_experts","type":"bool","label":"Offload expert weights","default":false},
|
||
{"name":"pin_experts","scope":"session","session_option":"ming3_audio.pin_experts","type":"bool","label":"Keep experts in pinned host RAM","default":false}
|
||
],
|
||
"xtts_v2": [
|
||
{"name":"temperature","type":"slider","label":"Temperature","default":0.75,"minimum":0,"maximum":2,"step":0.05},
|
||
{"name":"top_k","type":"number","label":"Top-k","default":50,"minimum":0,"step":1},
|
||
{"name":"top_p","type":"slider","label":"Top-p","default":0.85,"minimum":0,"maximum":1,"step":0.01},
|
||
{"name":"repetition_penalty","type":"number","label":"Repetition penalty","default":10,"minimum":1,"step":0.5},
|
||
{"name":"max_mel_tokens","type":"number","label":"Maximum mel tokens","default":605,"minimum":1,"step":1}
|
||
],
|
||
"latentsync": [
|
||
{"name":"num_inference_steps","type":"number","label":"Denoising steps","default":20,"minimum":1,"maximum":999,"step":1},
|
||
{"name":"cache_interval","type":"number","label":"DeepCache interval","default":1,"minimum":1,"step":1},
|
||
{"name":"guidance_scale","type":"number","label":"Guidance scale","default":1.5,"minimum":0,"step":0.1},
|
||
{"name":"frames_per_chunk","type":"number","label":"Frames per chunk","default":16,"minimum":1,"maximum":24,"step":1},
|
||
{"name":"fps","type":"number","label":"Frames per second","default":25,"minimum":1,"step":1},
|
||
{"name":"mem_saver","scope":"session","session_option":"latentsync.mem_saver","type":"bool","label":"Memory saver","default":false}
|
||
],
|
||
"ast_audioset": [
|
||
{"name":"top_k","type":"number","label":"Top classes","default":10,"minimum":1,"maximum":527,"step":1}
|
||
],
|
||
"ced": [
|
||
{"name":"top_k","type":"number","label":"Top classes","default":10,"minimum":1,"maximum":527,"step":1}
|
||
],
|
||
"fish_drama3": [
|
||
{"name":"text_chunk_size","type":"number","label":"Characters per chunk","default":128,"minimum":1,"step":1,"info":"Smaller chunks reduce waveform-decoder peak memory; larger chunks reduce repeated prefills."},
|
||
{"name":"temperature","type":"slider","label":"Temperature","default":1,"minimum":0,"maximum":2,"step":0.05},
|
||
{"name":"top_p","type":"slider","label":"Top-p","default":0.9,"minimum":0.01,"maximum":1,"step":0.01},
|
||
{"name":"top_k","type":"number","label":"Top-k","default":4097,"minimum":1,"step":1},
|
||
{"name":"disk_experts","type":"bool","scope":"session","session_option":"fish_drama3.disk_experts","label":"Offload expert weights","default":true,"info":"Keep only selected expert weights on the GPU. Expert computation stays on the selected backend; changing this reloads the model."},
|
||
{"name":"pin_experts","type":"bool","scope":"session","session_option":"fish_drama3.pin_experts","label":"Preload experts in pinned host RAM","default":true,"info":"Requires expert offload. Faster uploads, but holds the full expert banks in pinned host RAM."},
|
||
{"name":"expert_cache_size","type":"number","scope":"session","session_option":"fish_drama3.expert_cache_size","label":"GPU expert slots per layer","default":8,"minimum":8,"maximum":256,"step":1},
|
||
{"name":"context_length","type":"number","scope":"session","session_option":"fish_drama3.context_length","label":"Context capacity","default":4096,"minimum":1,"step":1,"info":"Prompt and generated frames share this capacity, including all voice references and dialogue history."}
|
||
],
|
||
"firered_vad": [
|
||
{"name":"threshold","type":"number","label":"Speech threshold","minimum":0,"maximum":1,"step":0.05},
|
||
{"name":"min_silence_duration_ms","type":"number","label":"Minimum silence (ms)","default":200,"minimum":0,"step":10},
|
||
{"name":"speech_pad_ms","type":"number","label":"Speech padding (ms)","default":0,"minimum":0,"step":10}
|
||
],
|
||
"hviske_asr_v6": [
|
||
{"name":"audio_chunk_mode","type":"choice","label":"Audio chunk mode","default":"auto","choices":["auto","silence","fixed","none"]},
|
||
{"name":"audio_chunk_duration_sec","type":"number","label":"Chunk duration (s)","default":28,"minimum":0.001,"maximum":30,"step":0.1},
|
||
{"name":"max_tokens","type":"number","label":"Maximum tokens","default":440,"minimum":1,"step":1},
|
||
{"name":"num_beams","type":"number","label":"Beam count","default":2,"minimum":1,"step":1},
|
||
{"name":"length_penalty","type":"number","label":"Length penalty","default":1,"minimum":0,"step":0.1}
|
||
],
|
||
"micro_wake_word": [
|
||
{"name":"threshold","type":"number","label":"Detection threshold","minimum":0,"maximum":1,"step":0.05},
|
||
{"name":"sliding_window_size","type":"number","label":"Sliding window frames","minimum":1,"step":1}
|
||
],
|
||
"open_wake_word": [
|
||
{"name":"threshold","type":"number","label":"Detection threshold","default":0.5,"minimum":0,"maximum":1,"step":0.05}
|
||
],
|
||
"plapre": [
|
||
{"name":"temperature","type":"number","label":"Temperature","default":0.8,"minimum":0,"step":0.05},
|
||
{"name":"top_p","type":"number","label":"Top-p","default":0.95,"minimum":0.01,"maximum":1,"step":0.01},
|
||
{"name":"top_k","type":"number","label":"Top-k","default":50,"minimum":-1,"step":1},
|
||
{"name":"x_vector_only_mode","type":"bool","label":"Reference speaker embedding only","default":false}
|
||
],
|
||
"mossformer2": [
|
||
{"name":"audio_chunk_duration_sec","type":"number","label":"Chunk duration (s, 0 = whole recording)","default":2,"minimum":0,"step":0.1},
|
||
{"name":"audio_chunk_overlap_sec","type":"number","label":"Chunk overlap (s)","default":0.5,"minimum":0,"step":0.1},
|
||
{"name":"normalize_output","type":"bool","label":"Normalize output","default":true}
|
||
],
|
||
"sherpa_kws": [
|
||
{"name":"keywords","type":"text","label":"Keyword tokens (one phrase per line)","lines":3,"placeholder":"Leave empty to use embedded keywords"},
|
||
{"name":"max_active_paths","type":"number","label":"Active decoding paths","default":4,"minimum":1,"step":1},
|
||
{"name":"keywords_score","type":"number","label":"Keyword score","default":1,"minimum":0,"step":0.1},
|
||
{"name":"keywords_threshold","type":"number","label":"Keyword threshold","default":0.25,"minimum":0,"maximum":1,"step":0.05},
|
||
{"name":"num_trailing_blanks","type":"number","label":"Trailing blanks","default":1,"minimum":0,"step":1}
|
||
],
|
||
"tf_gridnet": [
|
||
{"name":"audio_chunk_duration_sec","type":"number","label":"Chunk duration (s, 0 = whole recording)","default":2.4,"minimum":0,"step":0.1},
|
||
{"name":"audio_chunk_overlap_sec","type":"number","label":"Chunk overlap (s)","default":1.6,"minimum":0,"step":0.1},
|
||
{"name":"normalize_output","type":"bool","label":"Normalize output","default":true}
|
||
],
|
||
"voxtral_tts": [
|
||
{"name":"voice_id","type":"choice","label":"Voice preset","default":"casual_male","choices":["casual_female","casual_male","cheerful_female","neutral_female","neutral_male","pt_male","pt_female","nl_male","nl_female","it_male","it_female","fr_male","fr_female","es_male","es_female","de_male","de_female","ar_male","hi_male","hi_female"]},
|
||
{"name":"num_inference_steps","type":"number","label":"Diffusion steps","default":7,"minimum":1,"step":1},
|
||
{"name":"guidance_scale","type":"number","label":"Guidance scale","default":1.2,"minimum":0,"step":0.1},
|
||
{"name":"max_tokens","type":"number","label":"Maximum tokens","default":2048,"minimum":1,"step":1}
|
||
],
|
||
"kugelaudio": [
|
||
{"name":"num_inference_steps","type":"number","label":"Diffusion steps","default":20,"minimum":1,"maximum":1000,"step":1},
|
||
{"name":"guidance_scale","type":"number","label":"Guidance scale","default":3,"minimum":0,"step":0.1},
|
||
{"name":"max_tokens","type":"number","label":"Maximum tokens","default":2048,"minimum":1,"step":1},
|
||
{"name":"do_sample","type":"bool","label":"Sample AR tokens","default":false},
|
||
{"name":"temperature","type":"number","label":"Temperature","default":1,"minimum":0,"step":0.1},
|
||
{"name":"text_chunk_size","type":"number","label":"Text chunk size","default":300,"minimum":1,"step":1},
|
||
{"name":"text_chunk_mode","type":"choice","label":"Text chunk mode","default":"default","choices":["default","tag_aware","japanese","endline"]}
|
||
],
|
||
"crisperwhisper": [
|
||
{"name":"transcription_mode","type":"choice","label":"Transcription mode","default":"verbatim","choices":["verbatim","intended","verbatimize"]},
|
||
{"name":"return_timestamps","type":"bool","label":"Word timestamps","default":false},
|
||
{"name":"max_tokens","type":"number","label":"Maximum tokens","default":256,"minimum":1,"maximum":439,"step":1},
|
||
{"name":"audio_chunk_mode","type":"choice","label":"Audio chunk mode","default":"auto","choices":["auto","fixed","none"]},
|
||
{"name":"audio_chunk_duration_sec","type":"number","label":"Chunk duration (s)","default":30,"minimum":0.02,"maximum":30,"step":0.1},
|
||
{"name":"audio_chunk_overlap_sec","type":"number","label":"Chunk overlap (s)","default":4,"minimum":0,"step":0.1}
|
||
],
|
||
"index_echo": [
|
||
{"name":"target_language","type":"choice","label":"Translation language","default":"en","choices":["en","es","ja"]},
|
||
{"name":"max_text_tokens","type":"number","label":"Maximum text tokens","default":2000,"minimum":1,"step":1},
|
||
{"name":"temperature","type":"number","label":"Temperature","default":0,"minimum":0,"maximum":2,"step":0.1},
|
||
{"name":"glossary","type":"text","label":"Glossary","default":""},
|
||
{"name":"audio_chunk_mode","type":"choice","label":"Audio chunk mode","default":"auto","choices":["auto","quiet_energy","fixed","none"]},
|
||
{"name":"audio_chunk_duration_sec","type":"number","label":"Chunk duration (s)","default":60,"minimum":1,"maximum":60,"step":1}
|
||
],
|
||
"audio_flamingo": [
|
||
{"name":"instruct","type":"text","label":"Question or instruction","default":"Transcribe the input speech.","lines":3},
|
||
{"name":"max_tokens","type":"number","label":"Maximum tokens","default":2048,"minimum":1,"step":1},
|
||
{"name":"do_sample","type":"bool","label":"Sample tokens","default":false},
|
||
{"name":"temperature","type":"number","label":"Temperature","default":1,"minimum":0.01,"step":0.1},
|
||
{"name":"top_k","type":"number","label":"Top K","default":50,"minimum":0,"step":1},
|
||
{"name":"top_p","type":"number","label":"Top P","default":1,"minimum":0.01,"maximum":1,"step":0.05},
|
||
{"name":"repetition_penalty","type":"number","label":"Repetition penalty","default":1,"minimum":0.01,"step":0.05}
|
||
],
|
||
"owsm": [
|
||
{"name":"language","type":"text","label":"Input language (ISO 639-3)","default":"eng","placeholder":"eng, jpn, zho, auto"},
|
||
{"name":"target_language","type":"text","label":"Translation language (ISO 639-3)","default":""},
|
||
{"name":"return_timestamps","type":"bool","label":"Timestamp segments","default":false},
|
||
{"name":"condition_on_previous_text","type":"bool","label":"Condition on previous text","default":false},
|
||
{"name":"beam_size","type":"number","label":"Beam size","default":1,"minimum":1,"maximum":50002,"step":1},
|
||
{"name":"max_tokens","type":"number","label":"Maximum tokens (0 = model limit)","default":0,"minimum":0,"maximum":374,"step":1},
|
||
{"name":"audio_chunk_mode","type":"choice","label":"Audio chunk mode","default":"auto","choices":["auto","fixed","none"]},
|
||
{"name":"audio_chunk_duration_sec","type":"number","label":"Chunk duration (s)","default":30,"minimum":0.2,"maximum":30,"step":0.1}
|
||
],
|
||
"owsm_ctc": [
|
||
{"name":"language","type":"text","label":"Input language (ISO 639-3)","default":"eng","placeholder":"eng, jpn, zho, auto"},
|
||
{"name":"target_language","type":"text","label":"Translation language (ISO 639-3)","default":""},
|
||
{"name":"audio_chunk_mode","type":"choice","label":"Audio chunk mode","default":"auto","choices":["auto","fixed","none"]}
|
||
],
|
||
"reuse": [
|
||
{"name":"audio_chunk_duration_sec","type":"number","label":"Chunk duration (s, 0 = whole recording)","default":0,"minimum":0,"step":1},
|
||
{"name":"audio_chunk_overlap_sec","type":"number","label":"Chunk overlap (s)","default":1,"minimum":0,"step":0.1}
|
||
],
|
||
"smart_turn": [
|
||
{"name":"threshold","type":"number","label":"Completion threshold","default":0.5,"minimum":0,"maximum":1,"step":0.05}
|
||
],
|
||
"maya1": [
|
||
{"name":"instruct","type":"text","label":"Voice description","default":"Warm female voice with an American accent and conversational pacing.","lines":3},
|
||
{"name":"temperature","type":"number","label":"Temperature","default":0.4,"minimum":0.01,"step":0.05},
|
||
{"name":"top_p","type":"number","label":"Top P","default":0.9,"minimum":0,"maximum":1,"step":0.05},
|
||
{"name":"repetition_penalty","type":"number","label":"Repetition penalty","default":1.1,"minimum":0.01,"step":0.05}
|
||
],
|
||
"gigaam_asr": [
|
||
{"name":"return_timestamps","type":"bool","label":"Word timestamps","default":false},
|
||
{"name":"audio_chunk_mode","type":"choice","label":"Audio chunk mode","default":"auto","choices":["auto","fixed","quiet_energy","none"]},
|
||
{"name":"audio_chunk_duration_sec","type":"number","label":"Audio chunk duration (s)","default":25,"minimum":0.02,"maximum":25,"step":0.1}
|
||
],
|
||
"samsone": [
|
||
{"name":"instruct","type":"text","label":"Question or instruction","default":"describe the audio","lines":3},
|
||
{"name":"max_tokens","type":"number","label":"Maximum tokens","default":100,"minimum":1,"maximum":512,"step":1}
|
||
],
|
||
"sam_audio": [
|
||
{"name":"num_inference_steps","type":"number","label":"Inference steps","default":16,"minimum":1,"step":1},
|
||
{"name":"anchors","type":"text","label":"Time anchors (JSON)","default":"","placeholder":"[[\"+\",0.5,2.0],[\"-\",3.0,4.0]]"}
|
||
],
|
||
"tone_color_vc": [
|
||
{"name":"temperature","type":"number","label":"Temperature","default":0.3,"minimum":0,"step":0.05}
|
||
],
|
||
"moss_ttsd": [
|
||
{"name":"instruct","type":"text","label":"Delivery instruction","default":""},
|
||
{"name":"temperature","type":"number","label":"Temperature","default":1.5,"minimum":0,"step":0.1},
|
||
{"name":"top_p","type":"number","label":"Top P","default":0.6,"minimum":0,"maximum":1,"step":0.05},
|
||
{"name":"top_k","type":"number","label":"Top K","default":50,"minimum":0,"step":1},
|
||
{"name":"repetition_penalty","type":"number","label":"Repetition penalty","default":1.1,"minimum":0,"step":0.05},
|
||
{"name":"max_frames","type":"number","label":"Maximum audio frames (0 = automatic)","default":0,"minimum":0,"step":1}
|
||
],
|
||
"moss_voicegen": [
|
||
{"name":"temperature","type":"number","label":"Temperature","default":1.5,"minimum":0,"step":0.1},
|
||
{"name":"top_p","type":"number","label":"Top P","default":0.6,"minimum":0,"maximum":1,"step":0.05},
|
||
{"name":"top_k","type":"number","label":"Top K","default":50,"minimum":0,"step":1},
|
||
{"name":"repetition_penalty","type":"number","label":"Repetition penalty","default":1.1,"minimum":0,"step":0.05}
|
||
],
|
||
"canary_asr": [
|
||
{"name": "target_language", "type": "choice", "label": "Target language", "default": "", "choices": ["", "en", "de", "es", "fr"]},
|
||
{"name": "pnc", "type": "bool", "label": "Punctuation and capitalization", "default": true},
|
||
{"name": "audio_chunk_mode", "type": "choice", "label": "Audio chunk mode", "default": "auto", "choices": ["auto", "fixed", "none"]},
|
||
{"name": "audio_chunk_duration_sec", "type": "number", "label": "Audio chunk duration (s)", "default": 40, "minimum": 0.02, "maximum": 40, "step": 0.1}
|
||
],
|
||
"cohere_asr": [
|
||
{"name": "pnc", "type": "bool", "label": "Punctuation and capitalization", "default": true},
|
||
{"name": "audio_chunk_mode", "type": "choice", "label": "Audio chunk mode", "default": "auto", "choices": ["auto", "quiet_energy", "fixed", "none"]},
|
||
{"name": "audio_chunk_duration_sec", "type": "number", "label": "Audio chunk duration (s)", "default": 35, "minimum": 0.04, "maximum": 35, "step": 0.1}
|
||
],
|
||
"moss_transcribe_diarize": [
|
||
{"name": "instruct", "type": "text", "label": "Instruction", "default": "", "lines": 3}
|
||
],
|
||
"apollo": [
|
||
{"name": "audio_chunk_duration_sec", "type": "number", "label": "Audio chunk duration (s)", "default": 0, "minimum": 0, "step": 1},
|
||
{"name": "audio_chunk_overlap_sec", "type": "number", "label": "Audio chunk overlap (s)", "default": 1, "minimum": 0, "step": 0.1},
|
||
{"name": "edge_pad_duration_sec", "type": "number", "label": "Edge padding (s)", "default": 0, "minimum": 0, "step": 0.1}
|
||
],
|
||
"universr": [
|
||
{"name": "input_sample_rate", "type": "choice", "label": "Input bandwidth (Hz)", "default": "", "choices": ["", 8000, 12000, 16000, 24000]},
|
||
{"name": "audio_chunk_duration_sec", "type": "number", "label": "Audio chunk duration (s)", "default": 0, "minimum": 0, "step": 1},
|
||
{"name": "sampler_mode", "type": "choice", "label": "Sampler", "default": "midpoint", "choices": ["euler", "midpoint", "rk4"]},
|
||
{"name": "num_inference_steps", "type": "number", "label": "Inference steps", "default": 4, "minimum": 1, "step": 1},
|
||
{"name": "guidance_scale", "type": "number", "label": "Guidance scale", "default": 1.5, "minimum": 0, "step": 0.1}
|
||
],
|
||
"liveavatar": [
|
||
{"name": "reference_image_path", "type": "text", "label": "Reference image", "default": ""},
|
||
{"name": "generation_mode", "type": "choice", "label": "Generation mode", "default": "offline", "choices": ["offline", "liveavatar"]},
|
||
{"name": "height", "type": "number", "label": "Height", "default": 480, "minimum": 16, "step": 16, "precision": 0},
|
||
{"name": "width", "type": "number", "label": "Width", "default": 832, "minimum": 16, "step": 16, "precision": 0},
|
||
{"name": "num_frames", "type": "number", "label": "Frames per clip", "default": 48, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "num_clips", "type": "number", "label": "Maximum clips", "default": 1, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "num_inference_steps", "type": "number", "label": "Denoising steps", "default": 4, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "Guidance scale", "default": 0.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "shift", "type": "slider", "label": "Scheduler shift", "default": 3.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "negative_prompt", "type": "text", "label": "Negative prompt", "default": ""},
|
||
{"name": "fused_cfg", "type": "bool", "label": "Fused CFG", "default": true},
|
||
{"name": "sage_attention", "type": "bool", "label": "SageAttention", "default": true},
|
||
{"name": "memory_saver", "type": "bool", "label": "Memory saver", "default": true},
|
||
{"name": "denoiser_layerwise", "type": "bool", "label": "Layerwise denoiser", "default": false},
|
||
{"name": "denoiser_layerwise_batch", "type": "number", "label": "Layerwise batch", "default": 16, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "vae_encoder_chunk_size", "type": "number", "label": "VAE encoder chunk", "default": 16, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "vae_decoder_tile_size", "type": "number", "label": "VAE decoder tile", "default": 0, "minimum": 0, "step": 16, "precision": 0},
|
||
{"name": "target_cache_blocks", "type": "number", "label": "Target cache blocks", "default": 2, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "vae_cache_f16", "type": "bool", "label": "F16 VAE cache", "default": true},
|
||
{"name": "denoiser_weight_streaming", "type": "bool", "scope": "session", "session_option": "liveavatar.denoiser_weight_streaming", "label": "Stream denoiser weights", "default": false}
|
||
],
|
||
"pulsevad": [
|
||
{"name": "threshold", "type": "slider", "label": "Speech threshold", "default": 0.5, "minimum": 0, "maximum": 1, "step": 0.01},
|
||
{"name": "hop_size_samples", "type": "number", "label": "Hop size (samples)", "default": 1600, "minimum": 1, "step": 1},
|
||
{"name": "min_speech_duration_ms", "type": "number", "label": "Minimum speech (ms)", "default": 100, "minimum": 0, "step": 1},
|
||
{"name": "min_silence_duration_ms", "type": "number", "label": "Minimum silence (ms)", "default": 100, "minimum": 0, "step": 1}
|
||
],
|
||
"echo_tts": [
|
||
{"name": "num_inference_steps", "type": "slider", "label": "num_inference_steps", "label_en": "Sampling steps", "default": 40, "minimum": 8, "maximum": 40, "step": 1, "precision": 0, "info": "Euler sampler steps."},
|
||
{"name": "text_guidance_scale", "type": "slider", "label": "text_guidance_scale", "label_en": "Text guidance", "default": 3.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "speaker_guidance_scale", "type": "slider", "label": "speaker_guidance_scale", "label_en": "Speaker guidance", "default": 8.0, "minimum": 0.0, "maximum": 15.0, "step": 0.1},
|
||
{"name": "truncation_factor", "type": "slider", "label": "truncation_factor", "label_en": "Noise truncation", "default": 0.8, "minimum": 0.0, "maximum": 1.0, "step": 0.05},
|
||
{"name": "guidance_interval", "type": "slider", "label": "guidance_interval", "label_en": "Guidance interval", "default": 1, "minimum": 1, "maximum": 3, "step": 1, "precision": 0, "info": "Refresh the unconditional CFG lanes every Nth guided step. Higher is faster and works best with more steps; 1 is highest fidelity."},
|
||
{"name": "reference_duration_sec", "type": "slider", "label": "reference_duration_sec", "label_en": "Reference trim (s)", "default": 15.0, "minimum": 1.0, "maximum": 60.0, "step": 1.0, "info": "Trim the speaker reference before encoding. Around 10 s usually clones best."},
|
||
{"name": "seed", "type": "number", "label": "seed", "label_en": "Seed", "default": 0, "minimum": 0, "step": 1, "precision": 0}
|
||
],
|
||
"_comment": "WebUI TTS 高级参数控件配置：按模型 family 动态生成控件（gr.render）。每项字段：name=选项键；type=slider|number|bool|text|choice；scope=session 时写入 session_options，否则随请求 options 透传给模型；label/info=显示文案；default=默认值(应等于模型默认，已按 src/models/<family>/*.cpp 校对)；minimum/maximum/step=数值范围；precision=0 表示整数；choices=下拉候选。seed/max_tokens 已有专用输入框，勿在此重复；参考文本用『参考文本』框(reference_text)；文件路径/parity 类参数(如 *_noise_file)未纳入，可用『其它参数(JSON)』兜底框传。",
|
||
|
||
"qwen3_tts": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.9, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 50, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 1.0, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "repetition_penalty", "type": "slider", "label": "repetition_penalty", "default": 1.05, "minimum": 1.0, "maximum": 2.0, "step": 0.01},
|
||
{"name": "do_sample", "type": "bool", "label": "do_sample", "default": true},
|
||
{"name": "instruct", "type": "text", "label": "instruct（仅 VoiceDesign/CustomVoice）", "default": "", "placeholder": "风格/音色指令，Base 版忽略"},
|
||
{"name": "speaker", "type": "text", "label": "speaker（仅 CustomVoice）", "default": "", "placeholder": "内置音色名，其它版忽略"}
|
||
],
|
||
|
||
"vibevoice": [
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0, "info": "扩散步数（官方默认 10），越大越慢越稳"},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 1.3, "minimum": 0.0, "maximum": 5.0, "step": 0.1, "info": "CFG 引导强度"},
|
||
{"name": "max_length_times", "type": "number", "label": "max_length_times", "default": 2.0, "minimum": 0.1, "step": 0.1, "info": "最大输出长度倍数"},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 1.0, "minimum": 0.05, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 1.0, "minimum": 0.05, "maximum": 1.0, "step": 0.01},
|
||
{"name": "do_sample", "type": "bool", "label": "do_sample", "default": false},
|
||
{"name": "voice_samples", "type": "text", "label": "voice_samples（多说话人，逗号分隔 wav，≤4）", "default": "", "placeholder": "D:/a.wav,D:/b.wav — 用此项时勿再上传参考音色"}
|
||
],
|
||
|
||
"voxcpm2": [
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0, "info": "CFM/DiT 步数"},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 5.0, "step": 0.1},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "tag_aware", "choices": ["default", "tag_aware", "japanese", "endline"]},
|
||
{"name": "min_tokens", "type": "number", "label": "min_tokens", "default": 2, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "retry_badcase", "type": "bool", "label": "retry_badcase（自动重试异常输出）", "default": true}
|
||
],
|
||
|
||
"voxcpm1": [
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0, "info": "CFM/DiT 步数"},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 5.0, "step": 0.1},
|
||
{"name": "min_tokens", "type": "number", "label": "min_tokens", "default": 2, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "retry_badcase", "type": "bool", "label": "retry_badcase（自动重试异常输出）", "default": true}
|
||
],
|
||
|
||
"miotts": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.8, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 50, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 1.0, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "repetition_penalty", "type": "slider", "label": "repetition_penalty", "default": 1.0, "minimum": 1.0, "maximum": 1.5, "step": 0.01},
|
||
{"name": "best_of_n", "type": "number", "label": "best_of_n（候选数，>1 自动开启）", "default": 1, "minimum": 1, "maximum": 8, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"chatterbox": [
|
||
{"name": "exaggeration", "type": "slider", "label": "exaggeration", "default": 0.5, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info": "改动后需重新『加载模型』才生效"},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 0.5, "minimum": 0.0, "maximum": 1.0, "step": 0.05, "info": "改动后需重新『加载模型』才生效"},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.8, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "repetition_penalty", "type": "slider", "label": "repetition_penalty", "default": 1.2, "minimum": 1.0, "maximum": 2.0, "step": 0.01}
|
||
],
|
||
|
||
"chatterbox-vc": [
|
||
{"name": "s3gen_cfg_rate", "type": "slider", "label": "s3gen_cfg_rate（音色引导强度）", "label_en": "s3gen_cfg_rate (voice guidance)", "default": 0.7, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps（生成步数）", "label_en": "num_inference_steps", "default": 10, "minimum": 1, "maximum": 100, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"kokoro_tts": [
|
||
{"name": "speed", "type": "slider", "label": "speed（语速倍率）", "label_en": "speed (speech rate)", "default": 1.0, "minimum": 0.5, "maximum": 2.0, "step": 0.05}
|
||
],
|
||
|
||
"kitten_tts": [
|
||
{"name": "speed", "type": "slider", "label": "speed（语速倍率）", "label_en": "speed (speech rate)", "default": 1.0, "minimum": 0.01, "maximum": 2.0, "step": 0.05},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size（长文本分段字符数）", "label_en": "text_chunk_size", "default": 400, "minimum": 32, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "default", "choices": ["default", "tag_aware", "japanese", "endline"]}
|
||
],
|
||
|
||
"piper_tts": [
|
||
{"name": "speed", "type": "slider", "label": "speed（语速倍率）", "label_en": "speed (speech rate)", "default": 1.0, "minimum": 0.5, "maximum": 2.0, "step": 0.05},
|
||
{"name": "variation", "type": "slider", "label": "variation（音色变化）", "label_en": "variation", "default": 0.667, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "duration_variation", "type": "slider", "label": "duration_variation（时长变化）", "label_en": "duration_variation", "default": 0.8, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size（长文本分段字符数）", "label_en": "text_chunk_size", "default": 280, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "default", "choices": ["default", "tag_aware", "japanese", "endline"]}
|
||
],
|
||
|
||
"omnivoice": [
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 32, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 5.0, "step": 0.1},
|
||
{"name": "speed", "type": "slider", "label": "speed", "default": 1.0, "minimum": 0.5, "maximum": 2.0, "step": 0.05},
|
||
{"name": "instruct", "type": "text", "label": "instruct（风格/音色指令）", "default": "", "placeholder": "如：以轻快的语气朗读"}
|
||
],
|
||
|
||
"sense_asr": [
|
||
{"name": "enable_itn", "type": "bool", "label": "enable_itn（逆文本规范化）", "label_en": "enable_itn", "default": true},
|
||
{"name": "keep_tags", "type": "bool", "label": "keep_tags（保留语言/情绪/事件标签）", "label_en": "keep_tags", "default": false},
|
||
{"name": "audio_chunk_mode", "type": "choice", "label": "audio_chunk_mode", "default": "auto", "choices": ["auto", "fixed", "none"]},
|
||
{"name": "audio_chunk_duration_sec", "type": "number", "label": "audio_chunk_duration_sec", "default": 30, "minimum": 0.001, "step": 1}
|
||
],
|
||
|
||
"lfm2-audio-tts": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.8, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info_en": "Audio code sampling temperature; 0 is greedy."},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 64, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "default": 200, "minimum": 32, "step": 1, "precision": 0, "info_en": "Codepoints of text per spoken turn."}
|
||
],
|
||
|
||
"lfm2-audio-jp-tts": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.8, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info_en": "Audio code sampling temperature; 0 is greedy."},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 64, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "default": 200, "minimum": 32, "step": 1, "precision": 0, "info_en": "Codepoints of text per spoken turn."}
|
||
],
|
||
|
||
"lfm2-audio-s2s": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 1.0, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info_en": "Audio code sampling temperature; 0 is greedy."},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 4, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "text_temperature", "type": "slider", "label": "text_temperature", "default": 0.0, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info_en": "Text token sampling temperature; 0 picks the most likely token, as liquid-audio does."},
|
||
{"name": "text_top_k", "type": "number", "label": "text_top_k", "default": 0, "minimum": 0, "step": 1, "precision": 0, "info_en": "With text_temperature above 0, sample text tokens from the k most likely; 0 keeps all."}
|
||
],
|
||
|
||
"lfm2-audio-jp-s2s": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 1.0, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info_en": "Audio code sampling temperature; 0 is greedy."},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 4, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "text_temperature", "type": "slider", "label": "text_temperature", "default": 0.0, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info_en": "Text token sampling temperature; 0 picks the most likely token, as liquid-audio does."},
|
||
{"name": "text_top_k", "type": "number", "label": "text_top_k", "default": 0, "minimum": 0, "step": 1, "precision": 0, "info_en": "With text_temperature above 0, sample text tokens from the k most likely; 0 keeps all."}
|
||
],
|
||
|
||
"lfm2_audio": [
|
||
{"name": "audio_chunk_mode", "type": "choice", "label": "audio_chunk_mode", "default": "auto", "choices": ["auto", "vad", "fixed", "none"], "info_en": "auto keeps audio up to audio_chunk_seconds whole and splits longer audio at pauses; none transcribes it in one pass, which the model handles up to about a minute."},
|
||
{"name": "audio_chunk_seconds", "type": "number", "label": "audio_chunk_seconds", "default": 30, "minimum": 1, "step": 1},
|
||
{"name": "max_tokens", "type": "number", "label": "max_tokens", "default": 512, "minimum": 1, "step": 1, "precision": 0, "info_en": "Transcript tokens allowed per chunk; a transcript that reaches it is cut off there."}
|
||
],
|
||
|
||
"lfm3-audio-tts": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 1.0, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info_en": "Audio code temperature; 0 picks the most likely code."},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 128, "minimum": 0, "step": 1, "precision": 0, "info_en": "Sample audio codes from the k most likely; 0 keeps all."},
|
||
{"name": "repetition_penalty", "type": "number", "label": "repetition_penalty", "default": 1.1, "minimum": 0.01, "step": 0.05, "info_en": "Audio repetition penalty over the last 4096 frames of each codebook; must be positive, 1 disables it."},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "default": 200, "minimum": 1, "step": 1, "precision": 0, "info_en": "Target characters per text chunk; each chunk is its own spoken turn."}
|
||
],
|
||
|
||
"lfm3-audio-s2s": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 1.0, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info_en": "Audio code temperature; 0 picks the most likely code."},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 128, "minimum": 0, "step": 1, "precision": 0, "info_en": "Sample audio codes from the k most likely; 0 keeps all."},
|
||
{"name": "repetition_penalty", "type": "number", "label": "repetition_penalty", "default": 1.1, "minimum": 0.01, "step": 0.05, "info_en": "Audio repetition penalty over the last 4096 frames of each codebook; must be positive, 1 disables it."},
|
||
{"name": "text_temperature", "type": "slider", "label": "text_temperature", "default": 0.0, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info_en": "Text token temperature; 0 is greedy."},
|
||
{"name": "text_top_k", "type": "number", "label": "text_top_k", "default": 0, "minimum": 0, "step": 1, "precision": 0, "info_en": "Sample text tokens from the k most likely; 0 keeps all."},
|
||
{"name": "max_audio_tail_frames", "type": "number", "label": "max_audio_tail_frames", "default": 375, "minimum": 0, "step": 1, "precision": 0, "info_en": "Most audio frames after the reply text ends before the turn is closed with an end-of-audio frame; 0 disables the cap."}
|
||
],
|
||
|
||
"lfm3_audio": [
|
||
{"name": "audio_chunk_mode", "type": "choice", "label": "audio_chunk_mode", "default": "auto", "choices": ["auto", "vad", "fixed", "none"], "info_en": "auto keeps audio up to audio_chunk_duration_sec whole and splits longer audio at VAD pauses; none transcribes it in one pass."},
|
||
{"name": "audio_chunk_duration_sec", "type": "number", "label": "audio_chunk_duration_sec", "minimum": 1, "step": 1, "info_en": "Longest chunk in seconds; empty uses 300 or lfm3_audio.max_pass_duration_sec, whichever is lower."},
|
||
{"name": "chat", "type": "bool", "label": "chat", "default": false, "info_en": "Answer the spoken input as a text chat reply instead of transcribing it."},
|
||
{"name": "enable_thinking", "type": "bool", "label": "enable_thinking", "default": false, "info_en": "Needs chat. Let the model think first; only the text after </think> is returned."},
|
||
{"name": "max_tokens", "type": "number", "label": "max_tokens", "minimum": 1, "step": 1, "precision": 0, "info_en": "Text tokens allowed per chunk; empty uses 512 or 8 per second of audio, whichever is more (chat 512, 2048 with thinking). A reply that reaches it is cut off there."}
|
||
],
|
||
|
||
"confucius4_r2t2": [
|
||
{"name": "chunk_size_ms", "type": "slider", "scope": "session", "session_option": "confucius4_r2t2.chunk_size_ms", "label": "chunk_size_ms（流式分块毫秒）", "label_en": "chunk_size_ms (streaming chunk, ms)", "default": 320, "minimum": 80, "maximum": 2000, "step": 10, "precision": 0, "info": "80-2000ms：越小延迟越低；320ms 在 Apple Silicon 上延迟与速度较均衡。", "info_en": "80-2000 ms. Lower means lower latency; 320 ms balances latency and speed on Apple Silicon."},
|
||
{"name": "unfixed_chunk_num", "type": "number", "scope": "session", "session_option": "confucius4_r2t2.unfixed_chunk_num", "label": "unfixed_chunk_num（前 N 块不用稳定前缀）", "label_en": "unfixed_chunk_num (leading chunks without prefix)", "default": 2, "minimum": 0, "maximum": 10, "step": 1, "precision": 0, "info": "开头若干块不使用已识别文本作为前缀提示。", "info_en": "Leading chunks that decode without a stable-prefix prompt."},
|
||
{"name": "unfixed_token_num", "type": "number", "scope": "session", "session_option": "confucius4_r2t2.unfixed_token_num", "label": "unfixed_token_num（回滚 token 数）", "label_en": "unfixed_token_num (rollback tokens)", "default": 5, "minimum": 0, "maximum": 20, "step": 1, "precision": 0, "info": "作为前缀前从累积文本回滚的 token 数，用于降低边界抖动。", "info_en": "Tokens rolled back from the accumulated text before it is used as the prefix prompt."},
|
||
{"name": "rollback_punctuation", "type": "bool", "scope": "session", "session_option": "confucius4_r2t2.rollback_punctuation", "label": "rollback_punctuation（句末标点不回滚）", "label_en": "rollback_punctuation (keep trailing punctuation)", "default": false, "info": "输出已以标点结尾时不再回滚 token。", "info_en": "Do not roll back tokens when the output already ends with punctuation."},
|
||
{"name": "max_new_tokens", "type": "number", "scope": "session", "session_option": "confucius4_r2t2.max_tokens", "label": "max_new_tokens（每分块解码上限）", "label_en": "max_new_tokens (per-chunk decode budget)", "default": 32, "minimum": 1, "maximum": 256, "step": 1, "precision": 0, "info": "每个流式分块的贪婪解码上限（会话级，区别于离线 max_tokens）。", "info_en": "Greedy decode budget per streaming chunk (session-scoped; distinct from offline max_tokens)."}
|
||
],
|
||
|
||
"pocket_tts": [
|
||
{"name": "frames_after_eos", "type": "number", "label": "frames_after_eos（-1=自动）", "default": -1, "minimum": -1, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"gradium_phonon": [
|
||
{"name": "frames_after_eos", "type": "number", "label": "frames_after_eos（-1=自动）", "label_en": "frames_after_eos (-1 = auto)", "default": -1, "minimum": -1, "step": 1, "precision": 0},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.3, "minimum": 0.05, "maximum": 1.0, "step": 0.05}
|
||
],
|
||
|
||
"neutts": [
|
||
{"name": "voice_id", "type": "choice", "label": "voice_id（内置音色）", "label_en": "voice_id (built-in voice)", "default": "emily", "choices": ["dave", "emily", "greta", "jo", "juliette", "mateo", "paul", "sophie", "steven"]},
|
||
{"name": "emotion", "type": "choice", "label": "emotion（情绪）", "label_en": "emotion", "default": "neutral", "choices": ["angry", "disgusted", "sad", "happy", "fearful", "neutral", "surprised"]}
|
||
],
|
||
|
||
"magpie_tts": [
|
||
{"name": "language", "type": "choice", "label": "language", "default": "en", "choices": ["en", "ar-AE", "ar-MSA", "ar-SA", "de", "es", "fr", "hi", "it", "ko", "pt-BR", "vi", "zh"]},
|
||
{"name": "voice_id", "type": "choice", "label": "voice_id（打包音色）", "label_en": "voice_id (packaged voice)", "default": "Aria", "choices": ["Aria", "Jason", "John", "Leo", "Sofia"]},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.6, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 80, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.5, "minimum": 0.0, "maximum": 6.0, "step": 0.1},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "default": 300, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "default", "choices": ["default", "tag_aware", "japanese", "endline"]}
|
||
],
|
||
|
||
"breeze_tts": [
|
||
{"name": "instruction", "type": "text", "label": "instruction", "label_en": "Instruction", "default": "", "placeholder": "Describe the target voice for VoiceDesign."},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "default": 600, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "default", "choices": ["default", "tag_aware", "japanese", "endline"]},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 1.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.9, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "depth_temperature", "type": "slider", "label": "depth_temperature", "default": 0.9, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 50, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 1.0, "minimum": 0.0, "maximum": 1.0, "step": 0.01}
|
||
],
|
||
|
||
"breeze-tts": [
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "default": 600, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "default", "choices": ["default", "tag_aware", "japanese", "endline"]},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 1.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.9, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "depth_temperature", "type": "slider", "label": "depth_temperature", "default": 0.9, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 50, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 1.0, "minimum": 0.0, "maximum": 1.0, "step": 0.01}
|
||
],
|
||
|
||
"cosyvoice3": [
|
||
{"name": "template_name", "type": "choice", "label": "template_name", "default": "zero_shot", "choices": ["zero_shot", "cross_lingual", "instruct"]},
|
||
{"name": "instruction", "type": "text", "label": "instruction", "label_en": "Instruction", "default": "", "placeholder": "Used by template_name=instruct."},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "default": 600, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "default", "choices": ["default", "tag_aware", "japanese", "endline"]},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "min_tokens", "type": "number", "label": "min_tokens", "default": 0, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 25, "minimum": 1, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"inflect_v2": [
|
||
{"name": "speaking_rate", "type": "slider", "label": "speaking_rate（语速倍率）", "label_en": "speaking_rate", "default": 1.0, "minimum": 0.5, "maximum": 2.0, "step": 0.05},
|
||
{"name": "variation", "type": "slider", "label": "variation（音色变化）", "label_en": "variation", "default": 0.667, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size（长文本分段字符数）", "label_en": "text_chunk_size", "default": 280, "minimum": 1, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"sanotts": [
|
||
{"name": "speaking_rate", "type": "slider", "label": "speaking_rate（语速倍率）", "label_en": "speaking_rate", "default": 1.0, "minimum": 0.5, "maximum": 2.0, "step": 0.05},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size（长文本分段字符数）", "label_en": "text_chunk_size", "default": 280, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "word_budget", "choices": ["word_budget"]}
|
||
],
|
||
|
||
"dramabox": [
|
||
{"name": "negative_prompt", "type": "text", "label": "negative_prompt（负向提示）", "label_en": "negative_prompt", "default": "", "placeholder": "留空=模型内置质量提示", "placeholder_en": "Blank = built-in quality prompt"},
|
||
{"name": "duration_sec", "type": "number", "label": "duration_sec（0=自动估时）", "label_en": "duration_sec (0 = auto)", "default": 0.0, "minimum": 0.0, "step": 0.5},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 30, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.5, "minimum": 0.0, "maximum": 8.0, "step": 0.1},
|
||
{"name": "spatio_temporal_guidance_scale", "type": "slider", "label": "spatio_temporal_guidance_scale", "default": 1.5, "minimum": 0.0, "maximum": 5.0, "step": 0.1},
|
||
{"name": "duration_scale", "type": "slider", "label": "duration_scale（自动估时倍率）", "label_en": "duration_scale", "default": 1.1, "minimum": 0.5, "maximum": 2.0, "step": 0.05},
|
||
{"name": "reference_duration_sec", "type": "number", "label": "reference_duration_sec（参考音频裁剪/重复秒数）", "label_en": "reference_duration_sec", "default": 10.0, "minimum": 0.0, "step": 0.5},
|
||
{"name": "guidance_rescale", "type": "text", "label": "guidance_rescale", "default": "auto", "placeholder": "auto 或数值"},
|
||
{"name": "audio_chunk_threshold_sec", "type": "number", "label": "audio_chunk_threshold_sec（长文本阈值）", "label_en": "audio_chunk_threshold_sec", "default": 45.0, "minimum": 0.0, "step": 1.0},
|
||
{"name": "audio_chunk_duration_sec", "type": "number", "label": "audio_chunk_duration_sec（长文本分段目标时长）", "label_en": "audio_chunk_duration_sec", "default": 37.0, "minimum": 0.0, "step": 1.0},
|
||
{"name": "cross_fade_duration_sec", "type": "number", "label": "cross_fade_duration_sec（分段交叉淡化）", "label_en": "cross_fade_duration_sec", "default": 0.05, "minimum": 0.0, "step": 0.01}
|
||
],
|
||
|
||
"confucius4_tts": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.8, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 0.8, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 30, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "num_beams", "type": "number", "label": "num_beams", "default": 3, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "repetition_penalty", "type": "slider", "label": "repetition_penalty", "default": 10.0, "minimum": 0.0, "maximum": 20.0, "step": 0.1},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 25, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 0.7, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "default": 80, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "default", "choices": ["default", "tag_aware", "japanese", "endline"]},
|
||
{"name": "cross_fade_duration_sec", "type": "number", "label": "cross_fade_duration_sec", "default": 0.3, "minimum": 0.0, "step": 0.05},
|
||
{"name": "edge_fade_duration_sec", "type": "number", "label": "edge_fade_duration_sec", "default": 0.1, "minimum": 0.0, "step": 0.05},
|
||
{"name": "edge_pad_duration_sec", "type": "number", "label": "edge_pad_duration_sec", "default": 0.1, "minimum": 0.0, "step": 0.05}
|
||
],
|
||
|
||
"ace_step": [
|
||
{"name": "lora", "type": "text", "scope": "session", "session_option": "ace_step.lora", "label": "LoRA adapter", "default": "", "info": "Adapter file or PEFT directory on the server; relative paths use the model root."},
|
||
{"name": "lora_scale", "type": "number", "scope": "session", "session_option": "ace_step.lora_scale", "label": "LoRA strength", "default": 1, "step": 0.1, "info": "Multiplies the adapter metadata scale; zero disables the adapter."},
|
||
{"name": "route", "type": "choice", "label": "route（操作类型）", "default": "text2music", "choices": ["text2music", "complete", "lego", "extract", "cover", "cover-nofsq", "repaint", "remix"], "info": "cover/remix=换词翻唱，非 text2music 需上传源音频；详见 webui/README.md"},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 8, "minimum": 1, "maximum": 20, "step": 1, "precision": 0, "info": "扩散步数（turbo 上限 20）；remix 路由不填时默认 16，其他路由默认 8"},
|
||
{"name": "shift", "type": "slider", "label": "shift（时间步弯曲）", "default": 3.0, "minimum": 1.0, "maximum": 5.0, "step": 0.5, "info": "原版 turbo 默认 3.0；1.0 会明显劣化 remix 换词咬字"},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 1.0, "minimum": 0.0, "maximum": 5.0, "step": 0.1},
|
||
{"name": "audio_cover_strength", "type": "slider", "label": "【cover】audio_cover_strength", "default": 1.0, "minimum": 0.0, "maximum": 1.0, "step": 0.05, "info": "1=贴近原曲，0=自由发挥；建议 0.5"},
|
||
{"name": "cover_noise_strength", "type": "slider", "label": "【cover】cover_noise_strength", "default": 0.0, "minimum": 0.0, "maximum": 1.0, "step": 0.05, "info": "保旋律强度；推荐 0.1~0.25"},
|
||
{"name": "source_caption", "type": "text", "label": "【remix】source_caption", "default": "", "placeholder": "源歌曲描述；『🔍 分析』自动填"},
|
||
{"name": "source_lyrics", "type": "text", "lines": 4, "label": "【remix】source_lyrics", "default": "", "placeholder": "源歌曲原歌词；『🔍 分析』自动填"},
|
||
{"name": "flow_edit_n_min", "type": "slider", "label": "【remix】flow_edit_n_min", "default": 0.0, "minimum": 0.0, "maximum": 1.0, "step": 0.05, "info": "调大更保源曲、换词更弱"},
|
||
{"name": "flow_edit_n_max", "type": "slider", "label": "【remix】flow_edit_n_max", "default": 1.0, "minimum": 0.0, "maximum": 1.0, "step": 0.05, "info": "唱不出新歌词时降到 0.7~0.9"},
|
||
{"name": "flow_edit_n_avg", "type": "number", "label": "【remix】flow_edit_n_avg", "default": 2, "minimum": 1, "maximum": 4, "step": 1, "precision": 0, "info": "每步多次采样取平均（remix 默认 2）；1=最快"},
|
||
{"name": "bpm", "type": "number", "label": "【曲谱】BPM", "default": 0, "minimum": 0, "step": 1, "precision": 0, "info": "0=不指定"},
|
||
{"name": "keyscale", "type": "text", "label": "【曲谱】keyscale", "default": "", "placeholder": "如 F major"},
|
||
{"name": "timesignature", "type": "text", "label": "【曲谱】timesignature", "default": "", "placeholder": "如 4"}
|
||
],
|
||
|
||
"minimax_music3": [
|
||
{"name": "num_inference_steps", "type": "number", "label": "Flow steps per window", "default": 30, "minimum": 1, "maximum": 200, "step": 1, "precision": 0, "info": "Flow-matching Euler steps per 200-frame denoising window."},
|
||
{"name": "guidance_scale", "type": "slider", "label": "Flow guidance scale", "default": 1.7, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "ar_guidance_scale", "type": "slider", "label": "AR guidance scale", "default": 1.5, "minimum": 0.0, "maximum": 10.0, "step": 0.1, "info": "Classifier-free guidance of the semantic and residual code sampling."},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 50, "minimum": 1, "maximum": 1024, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"auk": [
|
||
{"name": "model_gguf", "type": "choice", "scope": "session", "session_option": "auk.model_gguf", "label": "model_gguf", "label_en": "Generator weights", "default": "auk-base-f32.gguf", "choices": ["auk-base-f32.gguf", "auk-base-f16.gguf", "auk-base-q8_0.gguf", "auk-flash-f32.gguf", "auk-flash-f16.gguf", "auk-flash-q8_0.gguf"], "info": "Selects Base or AuK-Flash automatically. Reload the model after changing this value."},
|
||
{"name": "qwen_gguf", "type": "choice", "scope": "session", "session_option": "auk.qwen_gguf", "label": "qwen_gguf", "label_en": "Qwen conditioner", "default": "qwen2.5-omni-3b-bf16.gguf", "choices": ["qwen2.5-omni-3b-bf16.gguf", "qwen2.5-omni-3b-q8_0.gguf"], "info": "Reload the model after changing this value."},
|
||
{"name": "vae_gguf", "type": "choice", "scope": "session", "session_option": "auk.vae_gguf", "label": "vae_gguf", "label_en": "VAE weights", "default": "auk-vae-f32.gguf", "choices": ["auk-vae-f32.gguf"], "info": "Reload the model after changing this value."},
|
||
{"name": "attention", "type": "choice", "scope": "session", "session_option": "auk.attention", "label": "attention", "label_en": "Attention", "default": "auto", "choices": ["auto", "flash", "eager"], "info": "Flow-model attention mode. Reload the model after changing this value."},
|
||
{"name": "mem_saver", "type": "bool", "scope": "session", "session_option": "auk.mem_saver", "label": "mem_saver", "label_en": "Memory saver", "default": false, "info": "Release Qwen and generation weights between stages. Warm requests reload weights and rebuild graphs."},
|
||
{"name": "instruct", "type": "text", "label": "instruct", "label_en": "Voice description", "default": "", "placeholder": "A warm male voice speaking clear English.", "info": "Optional for TTS only. Without it, text carries the complete upstream instruction."},
|
||
{"name": "duration_sec", "type": "number", "label": "duration_sec", "label_en": "Output duration (s)", "default": "", "minimum": 0.02, "step": 0.02, "info": "Required for speech generation without reference audio; otherwise the reference duration is used."},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "label_en": "Sampling steps", "default": 32, "minimum": 1, "step": 1, "precision": 0, "info": "Base Euler steps. AuK-Flash always uses its fixed four-step schedule."},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "label_en": "Guidance scale", "default": 2.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1, "info": "Base guidance strength. AuK-Flash disables classifier-free guidance."},
|
||
{"name": "sway_sampling_coef", "type": "number", "label": "sway_sampling_coef", "label_en": "Sway coefficient", "default": -1.0, "step": 0.1, "info": "Base sampling schedule sway. AuK-Flash uses its fixed time grid."}
|
||
],
|
||
|
||
"yue2": [
|
||
{"name": "ar_lora", "type": "text", "scope": "session", "session_option": "yue2.ar_lora", "label": "ar_lora", "label_en": "AR LoRA adapter", "default": "", "placeholder": "/path/to/ar_lora_inst_v3abc.safetensors", "info": "Unfused AR adapter for score and semantic planning; relative paths resolve against the model root. Reload the model after changing this value."},
|
||
{"name": "ar_lora_scale", "type": "number", "scope": "session", "session_option": "yue2.ar_lora_scale", "label": "ar_lora_scale", "label_en": "AR LoRA strength", "default": 1, "step": 0.1, "info": "Scales the AR adapter deltas; 0 disables it. Reload the model after changing this value."},
|
||
{"name": "nar_lora", "type": "text", "scope": "session", "session_option": "yue2.nar_lora", "label": "nar_lora", "label_en": "NAR LoRA adapter", "default": "", "placeholder": "/path/to/nar_lora_joint_v4.safetensors", "info": "Unfused NAR adapter for acoustic detail; relative paths resolve against the model root. Reload the model after changing this value."},
|
||
{"name": "nar_lora_scale", "type": "number", "scope": "session", "session_option": "yue2.nar_lora_scale", "label": "nar_lora_scale", "label_en": "NAR LoRA strength", "default": 1, "step": 0.1, "info": "Scales the LoRA deltas only; any full vae2llm/llm2vae projection replacements in the adapter stay at full strength. 0 disables the entire adapter. Reload the model after changing this value."},
|
||
{"name": "main_gguf", "type": "choice", "scope": "session", "session_option": "yue2.model_gguf", "label": "main_gguf", "label_en": "Main weights", "default": "yue2-3b-q8_0.gguf", "choices": ["yue2-3b-q8_0.gguf", "yue2-3b-q4_0.gguf", "yue2-3b-bf16.gguf"], "info": "Reload the model after changing this value."},
|
||
{"name": "vae_gguf", "type": "choice", "scope": "session", "session_option": "yue2.vae_gguf", "label": "vae_gguf", "label_en": "VAE weights", "default": "yue2-vae-f16.gguf", "choices": ["yue2-vae-f16.gguf", "yue2-vae-f32.gguf"], "info": "Reload the model after changing this value."},
|
||
{"name": "style", "type": "text", "label": "style", "label_en": "Style", "default": "English, indie pop, bright acoustic guitar, soft drums, warm lead vocal, polished demo mix", "placeholder": "English, city pop, groovy bass, synth, energetic vocal"},
|
||
{"name": "abc", "type": "text", "label": "abc", "label_en": "ABC score", "default": "", "placeholder": "Optional ABC notation. Use cot=melody or cot=full.", "lines": 4},
|
||
{"name": "abc_file", "type": "text", "label": "abc_file", "label_en": "ABC file path", "default": "", "placeholder": "/path/to/score.abc"},
|
||
{"name": "cot", "type": "choice", "label": "cot", "label_en": "Planning route", "default": "off", "choices": ["off", "melody", "full"], "info": "off = direct generation; melody/full use or generate ABC planning."},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "label_en": "Semantic guidance", "default": 1.01, "minimum": 0.0, "maximum": 5.0, "step": 0.01},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "label_en": "NAR steps", "default": 8, "minimum": 1, "maximum": 64, "step": 1, "precision": 0},
|
||
{"name": "abc_temperature", "type": "slider", "label": "abc_temperature", "label_en": "ABC temperature", "default": 0.7, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "abc_top_p", "type": "slider", "label": "abc_top_p", "label_en": "ABC top-p", "default": 0.9, "minimum": 0.01, "maximum": 1.0, "step": 0.01},
|
||
{"name": "abc_top_k", "type": "number", "label": "abc_top_k", "label_en": "ABC top-k", "default": 30, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "abc_repetition_penalty", "type": "slider", "label": "abc_repetition_penalty", "label_en": "ABC repetition penalty", "default": 1.005, "minimum": 0.1, "maximum": 2.0, "step": 0.001},
|
||
{"name": "abc_penalty_window", "type": "number", "label": "abc_penalty_window", "label_en": "ABC penalty window", "default": 100, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "abc_min_tokens", "type": "number", "label": "abc_min_tokens", "label_en": "ABC min tokens", "default": 32, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "abc_max_tokens", "type": "number", "label": "abc_max_tokens", "label_en": "ABC max tokens", "default": 4096, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "semantic_temperature", "type": "slider", "label": "semantic_temperature", "label_en": "Semantic temperature", "default": 1.0, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "semantic_top_p", "type": "slider", "label": "semantic_top_p", "label_en": "Semantic top-p", "default": 0.95, "minimum": 0.01, "maximum": 1.0, "step": 0.01},
|
||
{"name": "semantic_top_k", "type": "number", "label": "semantic_top_k", "label_en": "Semantic top-k", "default": 100, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "semantic_repetition_penalty", "type": "slider", "label": "semantic_repetition_penalty", "label_en": "Semantic repetition penalty", "default": 1.2, "minimum": 0.1, "maximum": 2.0, "step": 0.01},
|
||
{"name": "semantic_penalty_window", "type": "number", "label": "semantic_penalty_window", "label_en": "Semantic penalty window", "default": 50, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "semantic_min_tokens", "type": "number", "label": "semantic_min_tokens", "label_en": "Semantic min tokens", "default": 200, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "semantic_max_tokens", "type": "number", "label": "semantic_max_tokens", "label_en": "Semantic max tokens", "default": 9000, "minimum": 1, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"minimax_h3": [
|
||
{"name": "num_inference_steps", "type": "number", "label": "Denoising steps", "default": 12, "minimum": 1, "maximum": 50, "step": 1, "precision": 0, "info": "Twelve denoising steps provide a practical quality and performance balance."},
|
||
{"name": "num_frames", "type": "number", "label": "Output frames", "default": 241, "minimum": 5, "maximum": 1441, "step": 4, "precision": 0, "info": "Approximately 24 frames per output second; 241 frames produces about 10 seconds of audio."},
|
||
{"name": "guidance_scale", "type": "slider", "label": "Guidance scale", "default": 1.0, "minimum": 0.0, "maximum": 5.0, "step": 0.1},
|
||
{"name": "sampler", "type": "choice", "label": "Sampler", "default": "euler", "choices": ["euler", "res_multistep", "dpmpp_2m", "unipc"]},
|
||
{"name": "dit_acceleration", "type": "choice", "label": "DiT acceleration", "default": "none", "choices": ["none", "spectrum", "first_block_cache"], "info": "None uses the quality-first full-DiT path. Acceleration modes are experimental and may distort some outputs."},
|
||
{"name": "return_video", "type": "bool", "label": "Decode video", "default": false, "info": "Disabled by default to reduce memory use and return audio only."}
|
||
],
|
||
|
||
"stable_audio": [
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 8, "minimum": 1, "step": 1, "precision": 0, "info": "RF 扩散步数"},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 1.0, "minimum": 0.0, "maximum": 5.0, "step": 0.1},
|
||
{"name": "audio_input_kind", "type": "choice", "label": "audio_input_kind（仅上传源音频时生效）", "default": "init_audio", "choices": ["init_audio", "inpaint_audio"]},
|
||
{"name": "init_noise_level", "type": "slider", "label": "init_noise_level（init_audio 强度）", "default": 1.0, "minimum": 0.0, "maximum": 1.0, "step": 0.05}
|
||
],
|
||
|
||
"seed_vc": [
|
||
{"name": "route", "type": "choice", "label": "route（转换路径）", "default": "", "choices": ["", "v2_vc", "v1_whisper_bigvgan_vc", "v1_xlsr_hift_vc", "v1_svc"], "info": "留空=按任务默认"},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 30, "minimum": 1, "step": 1, "precision": 0, "info": "CFM 扩散步数"},
|
||
{"name": "length_adjust", "type": "slider", "label": "length_adjust（时长伸缩）", "default": 1.0, "minimum": 0.5, "maximum": 2.0, "step": 0.05},
|
||
{"name": "intelligibility_cfg_rate", "type": "slider", "label": "intelligibility_cfg_rate（仅 v2_vc）", "default": 0.7, "minimum": 0.0, "maximum": 1.0, "step": 0.05},
|
||
{"name": "similarity_cfg_rate", "type": "slider", "label": "similarity_cfg_rate（仅 v2_vc）", "default": 0.7, "minimum": 0.0, "maximum": 1.0, "step": 0.05},
|
||
{"name": "inference_cfg_rate", "type": "slider", "label": "inference_cfg_rate（仅 v1 路径）", "default": 0.7, "minimum": 0.0, "maximum": 1.0, "step": 0.05}
|
||
],
|
||
|
||
"rvc": [
|
||
{"name": "voice_id", "type": "choice", "label": "voice_id（打包音色）", "label_en": "voice_id", "default": "default", "choices": ["default", "manthos", "chocola", "fraise"]},
|
||
{"name": "voice_model_path", "type": "text", "label": "voice_model_path（自定义 RVC .pth/.pt）", "label_en": "voice_model_path", "default": "", "placeholder": "留空=使用打包音色"},
|
||
{"name": "retrieval_index_path", "type": "text", "label": "retrieval_index_path（FAISS index）", "label_en": "retrieval_index_path", "default": "", "placeholder": "可选 .index 路径"},
|
||
{"name": "retrieval_blend", "type": "slider", "label": "retrieval_blend", "default": 0.0, "minimum": 0.0, "maximum": 1.0, "step": 0.05},
|
||
{"name": "semitone_shift", "type": "number", "label": "semitone_shift（半音变调）", "label_en": "semitone_shift", "default": 0, "step": 1, "precision": 0},
|
||
{"name": "pitch_filter_radius", "type": "number", "label": "pitch_filter_radius", "default": 3, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "output_sample_rate", "type": "number", "label": "output_sample_rate（0=跟随音色）", "label_en": "output_sample_rate (0 = voice default)", "default": 0, "minimum": 0, "step": 1000, "precision": 0},
|
||
{"name": "rms_mix_rate", "type": "slider", "label": "rms_mix_rate", "default": 0.25, "minimum": 0.0, "maximum": 1.0, "step": 0.05},
|
||
{"name": "unvoiced_protection", "type": "slider", "label": "unvoiced_protection", "default": 0.33, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "speaker_id", "type": "number", "label": "speaker_id", "default": 0, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "audio_pad_duration_sec", "type": "number", "label": "audio_pad_duration_sec", "default": 1, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "split_query_sec", "type": "number", "label": "split_query_sec", "default": 5, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "split_center_sec", "type": "number", "label": "split_center_sec", "default": 30, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "split_threshold_sec", "type": "number", "label": "split_threshold_sec", "default": 32, "minimum": 1, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"meanvc2": [
|
||
{"name": "seed", "type": "number", "label": "seed", "default": 42, "minimum": 0, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"personaplex": [
|
||
{"name": "voice_id", "type": "choice", "label": "voice_id（打包音色）", "label_en": "voice_id (packaged voice)", "default": "NATF2", "choices": ["NATF0", "NATF1", "NATF2", "NATF3", "NATM0", "NATM1", "NATM2", "NATM3", "VARF0", "VARF1", "VARF2", "VARF3", "VARF4", "VARM0", "VARM1", "VARM2", "VARM3", "VARM4"]},
|
||
{"name": "system_prompt", "type": "text", "label": "system_prompt", "default": "", "placeholder": "Leave blank to use the text box as the system prompt."},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.8, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "text_temperature", "type": "slider", "label": "text_temperature", "default": 0.8, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 250, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "text_top_k", "type": "number", "label": "text_top_k", "default": 250, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "do_sample", "type": "bool", "label": "do_sample", "default": true}
|
||
],
|
||
|
||
"vevo2": [
|
||
{"name": "route", "type": "choice", "label": "route（任务路线）", "default": "", "choices": ["", "style_preserved_vc", "style_converted_vc", "style_preserved_svc", "style_converted_svc", "singing_style_conversion", "editing"], "info": "留空=按任务默认；详见 webui/README.md"},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 32, "minimum": 1, "step": 1, "precision": 0, "info": "流匹配步数"},
|
||
{"name": "use_pitch_shift", "type": "choice", "label": "use_pitch_shift（自动音高对齐）", "default": "", "choices": ["", "true", "false"], "info": "留空=按路线默认"},
|
||
{"name": "audio_chunk_duration_sec", "type": "number", "label": "audio_chunk_duration_sec（源音频分段秒数）", "default": 0.0, "minimum": 0.0, "step": 1.0, "info": "0=关闭；仅用于 source-audio VC/SVC 路线"},
|
||
{"name": "cross_fade_duration_sec", "type": "number", "label": "cross_fade_duration_sec（分段重叠/淡化秒数）", "default": 1.0, "minimum": 0.0, "step": 0.1, "info": "source-audio 分段启用时作为输入重叠和输出交叉淡化时长"},
|
||
{"name": "temperature", "type": "slider", "label": "temperature（AR 路线用）", "default": 0.7, "minimum": 0.0, "maximum": 2.0, "step": 0.05, "info": "默认取自模型 generation_config.json"},
|
||
{"name": "top_k", "type": "number", "label": "top_k（AR 路线用）", "default": 20, "minimum": 0, "step": 1, "precision": 0, "info": "默认取自模型 generation_config.json"},
|
||
{"name": "top_p", "type": "slider", "label": "top_p（AR 路线用）", "default": 0.8, "minimum": 0.0, "maximum": 1.0, "step": 0.01}
|
||
],
|
||
|
||
"heartmula": [
|
||
{"name": "tags", "type": "text", "label": "tags（逗号分隔）", "default": "pop", "placeholder": "pop,bright,drums,female vocals", "info": "风格/情绪/乐器/人声标签"},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 1.0, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 50, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale（MuLa CFG）", "default": 1.5, "minimum": 0.0, "maximum": 5.0, "step": 0.1},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps（codec 步数）", "default": 10, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "infinite_mode", "type": "bool", "label": "infinite_mode（长输出分段生成）", "default": false},
|
||
{"name": "codec_guidance_scale", "type": "slider", "label": "codec_guidance_scale", "default": 1.25, "minimum": 0.0, "maximum": 5.0, "step": 0.05}
|
||
],
|
||
|
||
"index_tts2": [
|
||
{"name": "lang", "type": "choice", "label": "lang（语种提示, 仅 IndexTTS2.5 模型）", "label_en": "lang (language hint, IndexTTS2.5 models only)", "default": "auto", "choices": ["auto", "zh", "en", "ja", "es", "ar"], "info": "仅对 IndexTTS2.5（多语种）模型生效：auto 含汉字按中文，否则按英文；日/西/阿建议显式选择", "info_en": "Only applies to IndexTTS2.5 (multilingual) models: auto picks zh when the text contains Han characters, otherwise en; set ja/es/ar explicitly"},
|
||
{"name": "emotion_text", "type": "text", "label": "emotion_text（情绪参考文本）", "label_en": "emotion_text (emotion reference text)", "default": "", "placeholder": "例：你吓死我了！你是鬼吗？", "placeholder_en": "e.g. You scared me to death!", "info": "填写后自动开启情感条件（use_emotion_text）", "info_en": "Setting this enables emotion conditioning."},
|
||
{"name": "emotion_alpha", "type": "slider", "label": "emotion_alpha（情感强度）", "label_en": "emotion_alpha", "default": 1.0, "minimum": 0.0, "maximum": 1.0, "step": 0.05},
|
||
{"name": "use_emotion_text", "type": "bool", "label": "use_emotion_text（从朗读文本推断情感）", "label_en": "use_emotion_text (infer from text)", "default": false},
|
||
{"name": "use_random_emotion", "type": "bool", "label": "use_random_emotion（随机情感）", "label_en": "use_random_emotion", "default": false},
|
||
{"name": "interval_silence_ms", "type": "number", "label": "interval_silence_ms（分段间静音）", "label_en": "interval_silence_ms", "default": 200, "minimum": 0, "step": 50, "precision": 0},
|
||
{"name": "duration_factor", "type": "slider", "label": "duration_factor（语速/时长倍率，>1 更慢，<1 更快）", "label_en": "duration_factor (duration multiplier; >1 slower, <1 faster)", "default": 1.0, "minimum": 0.5, "maximum": 2.0, "step": 0.05, "info": "对齐官方 IndexTTS2.5 的 duration_factor：缩放输出时长，不改变音色/内容", "info_en": "Matches official IndexTTS2.5 duration_factor: scales output duration without changing timbre or content"}
|
||
],
|
||
|
||
"zipvoice": [
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale（引导强度，0=关闭）", "label_en": "guidance_scale (0 disables CFG)", "default": 3.0, "minimum": 0.0, "maximum": 10.0, "step": 0.5},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps（Euler 步数）", "label_en": "num_inference_steps", "default": 8, "minimum": 1, "maximum": 64, "step": 1, "precision": 0},
|
||
{"name": "t_shift", "type": "slider", "label": "t_shift（时间步偏移，越小越偏低 SNR）", "label_en": "t_shift (timestep shift)", "default": 0.5, "minimum": 0.05, "maximum": 1.0, "step": 0.05},
|
||
{"name": "speed", "type": "slider", "label": "speed（语速倍率）", "label_en": "speed (duration multiplier)", "default": 1.0, "minimum": 0.5, "maximum": 2.0, "step": 0.05},
|
||
{"name": "lang", "type": "text", "label": "lang（英文段 espeak 语言）", "label_en": "lang (espeak voice for English runs)", "default": "en-us", "placeholder": "en-us"}
|
||
],
|
||
|
||
"irodori_tts": [
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps（RF 扩散步数）", "label_en": "num_inference_steps", "default": 40, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "duration_sec", "type": "number", "label": "duration_sec（0=模型自动预测时长）", "label_en": "duration_sec (0 = auto)", "default": 0, "minimum": 0, "step": 0.5},
|
||
{"name": "duration_scale", "type": "slider", "label": "duration_scale（语速倒数，越大越慢）", "label_en": "duration_scale", "default": 1.0, "minimum": 0.5, "maximum": 2.0, "step": 0.05}
|
||
],
|
||
|
||
"moss_tts_local": [
|
||
{"name": "do_sample", "type": "bool", "label": "do_sample", "default": true},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 1.7, "minimum": 0.0, "maximum": 2.5, "step": 0.05},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 0.8, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 25, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "repetition_penalty", "type": "slider", "label": "repetition_penalty", "default": 1.0, "minimum": 1.0, "maximum": 2.0, "step": 0.01}
|
||
],
|
||
|
||
"moss_tts_v15": [
|
||
{"name": "instruct", "type": "text", "label": "instruct（音色描述）", "label_en": "instruct (voice description)", "default": "", "placeholder": "参考音频比文字描述更可靠", "placeholder_en": "A reference recording is more reliable than a description."},
|
||
{"name": "language", "type": "choice", "label": "language（语言）", "label_en": "language", "default": "", "choices": ["", "English", "Chinese"]},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 1.5, "minimum": 0.0, "maximum": 3.0, "step": 0.05},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 0.6, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 50, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "repetition_penalty", "type": "slider", "label": "repetition_penalty", "default": 1.1, "minimum": 1.0, "maximum": 2.0, "step": 0.01}
|
||
],
|
||
|
||
"moss_tts_nano": [
|
||
{"name": "do_sample", "type": "bool", "label": "do_sample", "default": true},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 1.7, "minimum": 0.0, "maximum": 2.5, "step": 0.05},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 0.8, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 25, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "repetition_penalty", "type": "slider", "label": "repetition_penalty", "default": 1.0, "minimum": 1.0, "maximum": 2.0, "step": 0.01}
|
||
],
|
||
|
||
"audiosr": [
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 50, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 3.5, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "ddim_eta", "type": "slider", "label": "ddim_eta", "default": 1.0, "minimum": 0.0, "maximum": 1.0, "step": 0.05},
|
||
{"name": "audio_chunk_duration_sec", "type": "number", "label": "audio_chunk_duration_sec", "default": 15.0, "minimum": 1.0, "step": 1.0},
|
||
{"name": "audio_chunk_overlap_sec", "type": "number", "label": "audio_chunk_overlap_sec", "default": 2.0, "minimum": 0.0, "step": 0.5}
|
||
],
|
||
|
||
"controlfoley": [
|
||
{"name": "duration_sec", "type": "number", "label": "duration_sec", "default": 8.0, "minimum": 0.1, "step": 0.5},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 25, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 4.5, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "negative_prompt", "type": "text", "label": "negative_prompt", "default": "", "placeholder": "optional negative prompt"},
|
||
{"name": "mask_away_clip", "type": "bool", "label": "mask_away_clip", "default": false}
|
||
],
|
||
|
||
"midashenglm_gen": [
|
||
{"name": "duration_sec", "type": "number", "label": "duration_sec", "default": 20.0, "minimum": 0.1, "step": 0.5},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "stop_threshold", "type": "slider", "label": "stop_threshold", "default": 0.5, "minimum": 0.0, "maximum": 1.0, "step": 0.05},
|
||
{"name": "min_stop_step", "type": "number", "label": "min_stop_step", "default": 5, "minimum": 0, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"fireredtts3-base": [
|
||
{"name": "language", "type": "choice", "label": "language", "default": "Chinese", "choices": ["Chinese", "English", "Cantonese", "Japanese", "Korean", "Spanish", "French", "Russian", "Arabic", "Turkish", "Indonesian", "Portuguese", "Italian", "Dutch", "Vietnamese", "German", "Ukrainian", "Thai", "Polish", "Romanian", "Greek", "Czech", "Finnish", "Hindi", "ZH_Anhui", "ZH_Fujian", "ZH_Gansu", "ZH_Guizhou", "ZH_Hebei", "ZH_Henan", "ZH_Hubei", "ZH_Hunan", "ZH_Jiangxi", "ZH_Liaoning", "ZH_Minnan", "ZH_Ningxia", "ZH_Shaanxi", "ZH_Shandong", "ZH_Shanghai", "ZH_Shanxi", "ZH_Sichuan", "ZH_Tianjin", "ZH_Wenzhou", "ZH_Wu", "ZH_Yunnan"]},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "stop_threshold", "type": "slider", "label": "stop_threshold", "default": 0.5, "minimum": 0.0, "maximum": 1.0, "step": 0.05}
|
||
],
|
||
|
||
"fireredtts3-instruct": [
|
||
{"name": "template_name", "type": "choice", "label": "template_name", "default": "instruct_tts", "choices": ["instruct_tts"]},
|
||
{"name": "language", "type": "choice", "label": "language", "default": "Chinese", "choices": ["Chinese", "English", "Cantonese", "Japanese", "Korean", "Spanish", "French", "Russian", "Arabic", "Turkish", "Indonesian", "Portuguese", "Italian", "Dutch", "Vietnamese", "German", "Ukrainian", "Thai", "Polish", "Romanian", "Greek", "Czech", "Finnish", "Hindi", "ZH_Anhui", "ZH_Fujian", "ZH_Gansu", "ZH_Guizhou", "ZH_Hebei", "ZH_Henan", "ZH_Hubei", "ZH_Hunan", "ZH_Jiangxi", "ZH_Liaoning", "ZH_Minnan", "ZH_Ningxia", "ZH_Shaanxi", "ZH_Shandong", "ZH_Shanghai", "ZH_Shanxi", "ZH_Tianjin", "ZH_Wenzhou", "ZH_Wu", "ZH_Yunnan"]},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "stop_threshold", "type": "slider", "label": "stop_threshold", "default": 0.5, "minimum": 0.0, "maximum": 1.0, "step": 0.05},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "default": 600, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "text_chunk_mode", "type": "choice", "label": "text_chunk_mode", "default": "default", "choices": ["default", "tag_aware", "japanese", "endline"]}
|
||
],
|
||
|
||
"fireredtts3-instruct-vdesign": [
|
||
{"name": "template_name", "type": "choice", "label": "template_name", "default": "voice_design", "choices": ["voice_design"]},
|
||
{"name": "language", "type": "choice", "label": "language", "default": "Chinese", "choices": ["Chinese", "English", "Cantonese", "Japanese", "Korean", "Spanish", "French", "Russian", "Arabic", "Turkish", "Indonesian", "Portuguese", "Italian", "Dutch", "Vietnamese", "German", "Ukrainian", "Thai", "Polish", "Romanian", "Greek", "Czech", "Finnish", "Hindi", "ZH_Anhui", "ZH_Fujian", "ZH_Gansu", "ZH_Guizhou", "ZH_Hebei", "ZH_Henan", "ZH_Hubei", "ZH_Hunan", "ZH_Jiangxi", "ZH_Liaoning", "ZH_Minnan", "ZH_Ningxia", "ZH_Shaanxi", "ZH_Shandong", "ZH_Shanghai", "ZH_Shanxi", "ZH_Tianjin", "ZH_Wenzhou", "ZH_Wu", "ZH_Yunnan"]},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "stop_threshold", "type": "slider", "label": "stop_threshold", "default": 0.5, "minimum": 0.0, "maximum": 1.0, "step": 0.05}
|
||
],
|
||
|
||
"firered-audio-tts": [
|
||
{"name": "template_name", "type": "choice", "label": "template_name", "default": "tts_clone", "choices": ["tts_clone"]},
|
||
{"name": "language", "type": "choice", "label": "language", "default": "zh", "choices": ["zh", "en"]},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "max_new_audio_steps", "type": "number", "label": "max_new_audio_steps", "default": 750, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 20, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 0.8, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.7, "minimum": 0.0, "maximum": 2.0, "step": 0.05}
|
||
],
|
||
|
||
"firered-audio-vdesign": [
|
||
{"name": "template_name", "type": "choice", "label": "template_name", "default": "voice_design", "choices": ["voice_design"]},
|
||
{"name": "language", "type": "choice", "label": "language", "default": "zh", "choices": ["zh", "en"]},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "max_new_audio_steps", "type": "number", "label": "max_new_audio_steps", "default": 750, "minimum": 1, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"firered-audio-semantic-edit": [
|
||
{"name": "template_name", "type": "choice", "label": "template_name", "default": "semantic_edit", "choices": ["semantic_edit"]},
|
||
{"name": "language", "type": "choice", "label": "language", "default": "zh", "choices": ["zh", "en"]},
|
||
{"name": "instruction", "type": "text", "label": "instruction", "default": "", "placeholder": "delete '比普通的茶叶要'"},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "max_new_audio_steps", "type": "number", "label": "max_new_audio_steps", "default": 750, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "max_new_text_tokens", "type": "number", "label": "max_new_text_tokens", "default": 512, "minimum": 1, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"firered-audio-acoustic-edit": [
|
||
{"name": "template_name", "type": "choice", "label": "template_name", "default": "acoustic_edit", "choices": ["acoustic_edit"]},
|
||
{"name": "language", "type": "choice", "label": "language", "default": "zh", "choices": ["zh", "en"]},
|
||
{"name": "instruction", "type": "text", "label": "instruction", "default": "shift the pitch by 3 steps", "placeholder": "shift the pitch by 3 steps"},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "default": 10, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "guidance_scale", "type": "slider", "label": "guidance_scale", "default": 2.0, "minimum": 0.0, "maximum": 10.0, "step": 0.1},
|
||
{"name": "max_new_audio_steps", "type": "number", "label": "max_new_audio_steps", "default": 750, "minimum": 1, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"firered-audio-asr": [
|
||
{"name": "template_name", "type": "choice", "label": "template_name", "default": "asr", "choices": ["asr", "understand"]},
|
||
{"name": "language", "type": "choice", "label": "language", "default": "zh", "choices": ["zh", "en"]},
|
||
{"name": "enable_thinking", "type": "bool", "label": "enable_thinking", "default": false},
|
||
{"name": "max_new_tokens", "type": "number", "label": "max_new_tokens", "default": 512, "minimum": 1, "step": 1, "precision": 0},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 20, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 0.8, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.7, "minimum": 0.0, "maximum": 2.0, "step": 0.05}
|
||
],
|
||
|
||
"sopro_tts": [
|
||
{"name": "language", "type": "choice", "label": "language", "label_en": "Language tag", "default": "", "choices": ["", "en", "pt", "fr", "de"], "info": "Optional <|lang_xx|> tag; helps pronunciation on ambiguous text."},
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "label_en": "Temperature", "default": 0.8, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "label_en": "Top-p", "default": 0.9, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "label_en": "Top-k", "default": 25, "minimum": 0, "step": 1, "precision": 0, "info": "0 disables top-k truncation."},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps", "label_en": "Acoustic steps", "default": 2, "minimum": 1, "maximum": 32, "step": 1, "precision": 0, "info": "Rectified-flow Euler steps for the acoustic head."},
|
||
{"name": "max_seconds", "type": "number", "label": "max_seconds", "label_en": "Max seconds per segment", "default": 30.0, "minimum": 1.0, "maximum": 60.0, "step": 0.5, "precision": 1},
|
||
{"name": "min_seconds", "type": "number", "label": "min_seconds", "label_en": "Min seconds per segment", "default": 0.4, "minimum": 0.0, "maximum": 10.0, "step": 0.1, "precision": 1, "info": "Must not exceed max_seconds."},
|
||
{"name": "ref_seconds", "type": "number", "label": "ref_seconds", "label_en": "Reference seconds", "default": 10.0, "minimum": 1.0, "maximum": 30.0, "step": 0.5, "precision": 1, "info": "Reference window used for cloning."},
|
||
{"name": "text_chunk_size", "type": "number", "label": "text_chunk_size", "label_en": "Segment size", "default": 300, "minimum": 20, "maximum": 2000, "step": 10, "precision": 0, "info": "Max codepoints per synthesis segment."}
|
||
],
|
||
"supertonic": [
|
||
{"name": "voice", "type": "choice", "label": "voice（预置音色：M 男声 / F 女声）", "label_en": "voice (M = male, F = female presets)", "default": "M1", "choices": ["M1", "M2", "M3", "M4", "M5", "F1", "F2", "F3", "F4", "F5"]},
|
||
{"name": "speaking_rate", "type": "slider", "label": "speaking_rate（语速倍率）", "label_en": "speaking_rate", "default": 1.05, "minimum": 0.5, "maximum": 2.0, "step": 0.05},
|
||
{"name": "num_inference_steps", "type": "number", "label": "num_inference_steps（流匹配步数）", "label_en": "num_inference_steps", "default": 8, "minimum": 1, "step": 1, "precision": 0}
|
||
],
|
||
|
||
"audio8_tts": [
|
||
{"name": "temperature", "type": "slider", "label": "temperature", "default": 0.7, "minimum": 0.0, "maximum": 2.0, "step": 0.05},
|
||
{"name": "top_p", "type": "slider", "label": "top_p", "default": 0.9, "minimum": 0.0, "maximum": 1.0, "step": 0.01},
|
||
{"name": "top_k", "type": "number", "label": "top_k", "default": 50, "minimum": 0, "step": 1, "precision": 0},
|
||
{"name": "max_tokens", "type": "number", "label": "max_tokens", "default": 1024, "minimum": 1, "step": 1, "precision": 0}
|
||
]
|
||
}
|