voice_audio_multi/model_specs/owsm_ctc.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

303 lines
4.8 KiB
JSON

{
"family": "owsm_ctc",
"schema_version": 1,
"display_name": "OWSM-CTC",
"description": "ESPnet non-autoregressive multilingual speech recognition and speech translation with self-conditioned CTC.",
"category": "asr",
"status": "supported",
"tasks": [
"asr"
],
"modes": [
"offline",
"streaming"
],
"languages": [
"abk",
"afr",
"amh",
"ara",
"asm",
"ast",
"aze",
"bak",
"bas",
"bel",
"ben",
"bos",
"bre",
"bul",
"cat",
"ceb",
"ces",
"chv",
"ckb",
"cmn",
"cnh",
"cym",
"dan",
"deu",
"dgd",
"div",
"ell",
"eng",
"epo",
"est",
"eus",
"fas",
"fil",
"fin",
"fra",
"frr",
"ful",
"gle",
"glg",
"grn",
"guj",
"hat",
"hau",
"heb",
"hin",
"hrv",
"hsb",
"hun",
"hye",
"ibo",
"ina",
"ind",
"isl",
"ita",
"jav",
"jpn",
"kab",
"kam",
"kan",
"kat",
"kaz",
"kea",
"khm",
"kin",
"kir",
"kmr",
"kor",
"lao",
"lav",
"lga",
"lin",
"lit",
"ltz",
"lug",
"luo",
"mal",
"mar",
"mas",
"mdf",
"mhr",
"mkd",
"mlt",
"mon",
"mri",
"mrj",
"mya",
"myv",
"nan",
"nep",
"nld",
"nno",
"nob",
"npi",
"nso",
"nya",
"oci",
"ori",
"orm",
"ory",
"pan",
"pol",
"por",
"pus",
"quy",
"roh",
"ron",
"rus",
"sah",
"sat",
"sin",
"skr",
"slk",
"slv",
"sna",
"snd",
"som",
"sot",
"spa",
"srd",
"srp",
"sun",
"swa",
"swe",
"swh",
"tam",
"tat",
"tel",
"tgk",
"tgl",
"tha",
"tig",
"tir",
"tok",
"tpi",
"tsn",
"tuk",
"tur",
"twi",
"uig",
"ukr",
"umb",
"urd",
"uzb",
"vie",
"vot",
"wol",
"xho",
"yor",
"yue",
"zho",
"zul"
],
"capabilities": {},
"dependencies": [],
"options": {
"request": [
{
"name": "language",
"type": "string",
"description": "Input ISO 639-3 language code, such as eng, jpn, or zho. auto uses unknown-language conditioning.",
"required": false,
"default": "eng"
},
{
"name": "target_language",
"type": "string",
"description": "Optional OWSM translation target code. Omit for transcription.",
"required": false
},
{
"name": "audio_chunk_mode",
"type": "enum",
"description": "Long-form audio chunking mode.",
"values": [
"auto",
"fixed",
"none"
],
"required": false,
"default": "auto"
}
],
"session": [
{
"name": "weight_type",
"type": "enum",
"description": "Matmul weight storage override.",
"values": [
"native",
"f32",
"f16",
"bf16",
"q8_0",
"q4_0",
"q4_1",
"q5_0",
"q5_1",
"q2_k",
"q3_k",
"q4_k",
"q5_k",
"q6_k"
],
"required": false,
"default": "native"
}
],
"load": []
},
"runtime": {
"tags": [
"gguf"
]
},
"ui": {
"tags": [
"ASR",
"GGUF"
],
"docs": [
"docs/models/owsm_ctc.md"
],
"recommended_package": "owsm_ctc_v4_1b_f32"
},
"package_defaults": {
"download": {
"kind": "huggingface_snapshot",
"repo": "audio-cpp/audio.cpp-gguf",
"revision": "main",
"gated": false
}
},
"packages": [
{
"id": "owsm_ctc_v4_1b_f32",
"display_name": "OWSM-CTC v4 1B F32",
"default": true,
"format": "gguf",
"precision": "f32",
"target_directory": "OWSM-CTC-GGUF",
"files": [
"OWSM-CTC-GGUF/owsm-ctc-v4-1b-f32.gguf"
],
"strip_prefix": "OWSM-CTC-GGUF"
},
{
"id": "owsm_ctc_v4_1b_q8_0",
"display_name": "OWSM-CTC v4 1B Q8_0",
"default": false,
"format": "gguf",
"precision": "q8_0",
"target_directory": "OWSM-CTC-GGUF",
"files": [
"OWSM-CTC-GGUF/owsm-ctc-v4-1b-q8_0.gguf"
],
"strip_prefix": "OWSM-CTC-GGUF"
}
],
"sources": [
{
"format": "safetensors",
"roots": {
"model": "."
},
"files": {
"config": "model:config.json",
"tokenizer": "model:bpe.model"
},
"tensors": {
"weights": "model:model.safetensors"
}
},
{
"format": "gguf",
"roots": {
"model": ".",
"weights": "$gguf"
},
"files": {
"config": "model:config.json",
"tokenizer": "model:bpe.model"
},
"tensors": {
"weights": "weights:"
}
}
]
}