ModelListedExternal

Whisper Large v3 Onnx

ONNX export of [openai/whisper-large-v3][base] in the [transformers.js][tjs] layout, with the cross-attention alignment heads baked into generationconfig.json so the decoder emits word-level timestamps...

Sourcehuggingface.co/musetric/whisper-large-v3-onnx/tree/228b20358cc86d1f8cc30299dc940438a3a7e4a0

KindText modelBase modelopenai/whisper-large-v3VersionCgrade

Model source

Kind: Text model
File size: 762 MB
Base model: openai/whisper-large-v3
Version: vQ4
Updated: Sep 26, 2026
Downloads: 11
Source: Hugging Face

Source description

Whisper large-v3 — word-timestamped ONNX (q4 / WebGPU)

ONNX export of openai/whisper-large-v3 in the transformers.js layout, with the cross-attention alignment heads baked into generation_config.json so the decoder emits word-level timestamps (return_timestamps: 'word'). Packaged for the musetric packages/ai runtime (@huggingface/transformers on the WebGPU execution provider).

onnx-community publishes such _timestamped variants for turbo, small and tiny, but not for large-v3 — hence this export. It is produced by scripts/onnx/whisper in musetric-toolkit (optimum + the transformers.js converter).

Files

file	role
`encoder_model_q4.onnx`	audio encoder (q4)
`decoder_model_merged_q4.onnx`	decoder with merged KV-cache (q4)
`config.json`, `generation_config.json`, `preprocessor_config.json`	model / feature-extractor config
`tokenizer.json`, `tokenizer_config.json`, `vocab.json`, `merges.txt`, `added_tokens.json`, `special_tokens_map.json`, `normalizer.json`	tokenizer

Only the q4 graphs are shipped; the runtime loads dtype: 'q4' for both the encoder and the merged decoder.

How to use

import { pipeline } from '@huggingface/transformers';

const asr = await pipeline(
  'automatic-speech-recognition',
  'musetric/whisper-large-v3-onnx',
  {
    device: 'webgpu',
    subfolder: '',
    dtype: { encoder_model: 'q4', decoder_model_merged: 'q4' },
  },
);
const out = await asr(audio, { return_timestamps: 'word', chunk_length_s: 30 });

Intended uses & limitations

Intended: client/edge speech-to-text with word timings via WebGPU through @huggingface/transformers — e.g. lyric alignment in musetric.

Limitations:

Requires a WebGPU adapter with shader-f16.
q4 weight quantization trades a little accuracy for size/speed; for non-English audio quality varies (the musetric pipeline adds language forcing and loop guards on top).
Word timestamps come from the whisper cross-attention heads, not a separate forced aligner.

Source & lineage

Documented only as far as it is verifiable.

Base weights: openai/whisper-large-v3 @ 06f233fe06e710322aca913c1bc4249a0d71fce1. Inference-only re-export in the transformers.js layout; no fine-tuning.
Alignment heads: the [layer, head] pairs correlated with word timing come from hollance's gist, which covers tiny through large-v2. large-v3 is absent there, so its heads are taken from the generation_config.json of the base model itself.
Converter: the transformers.js converter plus optimum, pinned to transformers==4.42.4 / optimum==1.21.3 / torch==2.4.1 (transformers 4.43 adds a cache_position decoder input that @huggingface/transformers 4.2.0 does not feed).
Export tooling: scripts/onnx/whisper in musetric-toolkit.

License

Apache-2.0, following the model card the weights are downloaded from.

Upstream is inconsistent here and it is worth knowing: the openai/whisper repository states that "Whisper's code and model weights are released under the MIT License", while the Hugging Face model card these weights are actually fetched from declares apache-2.0. This export follows the source it downloads from; consult both before relying on either.

Sources

1 source

Hugging FacePrimaryVersion 228b20358cc86d1f8cc30299dc940438a3a7e4a0

huggingface.co/musetric/whisper-large-v3-onnx/tree/228b20358cc86d1f8cc30299dc940438a3a7e4a0

Verified Sep 6

Model artifacts

1 artifact

decoder_model_merged_q4.onnx

onnx · Q4 (unverified) · 762 MB · SHA-256 7b68ff483329…2e2b · Hugging Face

Download file

Source excerpts

2 excerpts

ReadmeSep 6huggingface.co/musetric/whisper-large-v3-onnx

--- license: apache-2.0 library_name: transformers.js pipeline_tag: automatic-speech-recognition base_model: openai/whisper-large-v3 tags: - audio - automatic-speech-recognition - whisper - onnx - transformers.js - webgpu --- # Whisper large-v3 — word-timestamped ONNX (q4 / WebGPU) ONNX export of [**openai/whisper-large-v3**][base] in the [transformers.js][tjs] layout, with the **cross-attention alignment heads** baked into `generation_config.json` so the decoder emits **word-level timestamps** (`return_timestamps: 'word'`). Packaged for the [`musetric`][musetric] `packages/ai` runtime ([`@huggingface/transformers`][tjs] on the **WebGPU** execution provider). [onnx-community][oc] publishes such `_timestamped` variants for [turbo][oc-turbo], [small][oc-small] and [tiny][oc-tiny], but not for **large-v3** — hence this export. It is produced by [`scripts/onnx/whisper`][sc] in [musetric-toolkit][toolkit] (optimum + the [transformers.js converter][conv]). ## Files | file | role | |---|---| | `encoder_model_q4.onnx` | audio encoder (q4) | | `decoder_model_merged_q4.onnx` | decoder with merged KV-cache (q4) | | `config.json`, `generation_config.json`, `preprocessor_config.json` | model / feature-extractor config | | `tokenizer.json`, `tokenizer_config.json`, `vocab.json`, `merges.txt`, `added_tokens.json`, `special_tokens_map.json`, `normalizer.json` | tokenizer | Only the **q4** graphs are shipped; the runtime loads `dtype: 'q4'` for both the encoder and the merged decoder. ## How to use ```ts import { pipeline } from '@huggingface/transformers'; const asr = await pipeline( 'automatic-speech-recognition', 'musetric/whisper-large-v3-onnx', { device: 'webgpu', subfolder: '', dtype: { encoder_model: 'q4', decoder_model_merged: 'q4' }, }, ); const out = await asr(audio, { return_timestamps: 'word', chunk_length_s: 30 }); ``` ## Intended uses & limitations **Intended:** client/edge speech-to-text with word timings via WebGPU through [`@huggingface/transformers`][tjs] — e.g. lyric alignment in `musetric`. **Limitations:** - Requires a WebGPU adapter with `shader-f16`. - q4 weight quantization trades a little accuracy for size/speed; for non-English audio quality varies (the `musetric` pipeline adds language forcing and loop guards on top). - Word timestamps come from the whisper cross-attention heads, not a separate forced aligner. ## Sou...

NoteSep 6huggingface.co/musetric/whisper-large-v3-onnx

Source context: 11 downloads · 0 likes · Pipeline automatic-speech-recognition · Library transformers.js · Repo musetric/whisper-large-v3-onnx