voice_audio_multi/model_specs/whistle_asr.json
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
Initial import: audio.cpp with Orange Pi documentation
2026-11-04 19:05:22 +01:00

117 lines
2.8 KiB
JSON

{
"schema_version": 1,
"family": "whistle_asr",
"display_name": "Whistle ASR",
"description": "Whistle offline speech recognition with FP32 GGUF weights, a GGML encoder and a host decoder.",
"category": "asr",
"status": "experimental",
"tasks": [
"asr"
],
"modes": [
"offline"
],
"languages": [
"en",
"de",
"fr",
"es",
"it",
"nl",
"pl"
],
"runtime": {
"tags": [
"gguf"
]
},
"capabilities": {},
"options": {
"request": [
{
"name": "language",
"type": "string",
"description": "Force one supported language; omit to detect the language.",
"required": false
}
],
"session": [
{
"name": "exec_plan",
"type": "string",
"description": "JSON component execution plans. Component decoder accepts impl native/neon (default native), both on CPU independently of the encoder backend. The experimental neon implementation accelerates the vocabulary projection with the framework NEON library and requires ENGINE_ENABLE_NEON=ON and AArch64 NEON. No artifacts or params are used.",
"required": false,
"default": "{}"
}
],
"load": []
},
"package_defaults": {
"download": {
"kind": "huggingface_snapshot",
"repo": "audio-cpp/audio.cpp-gguf",
"revision": "main",
"gated": false
}
},
"packages": [
{
"id": "whistle_asr_f32",
"display_name": "Whistle ASR GGUF F32",
"default": true,
"format": "gguf",
"precision": "f32",
"target_directory": "Whistle-GGUF",
"files": [
"Whistle-GGUF/whistle-f32.gguf"
],
"strip_prefix": "Whistle-GGUF"
}
],
"dependencies": [],
"ui": {
"tags": [
"ASR",
"GGUF"
],
"docs": [
"docs/models/whistle_asr.md"
],
"recommended_package": "whistle_asr_f32"
},
"sources": [
{
"format": "gguf",
"roots": {
"model": ".",
"weights": "$gguf"
},
"files": {
"config": "model:config.json",
"tokenizer_blob": "model:tokenizer.blob",
"mel_filterbank": "model:mel_filterbank_80.f32",
"hadamard_perm1": "model:hadamard_perm1.f32",
"hadamard_perm2": "model:hadamard_perm2.f32"
},
"tensors": {
"weights": "weights:"
}
},
{
"format": "safetensors",
"roots": {
"model": "."
},
"files": {
"config": "model:config.json",
"tokenizer_blob": "model:tokenizer.blob",
"mel_filterbank": "model:mel_filterbank_80.f32",
"hadamard_perm1": "model:hadamard_perm1.f32",
"hadamard_perm2": "model:hadamard_perm2.f32"
},
"tensors": {
"weights": "model:whistle.safetensors"
}
}
]
}