| --- |
| license: cc-by-4.0 |
| base_model: BUT-FIT/DiCoW_v3_3 |
| tags: [onnx, whisper, speaker-diarization, target-speaker-asr, dicow] |
| --- |
| |
| # DiCoW v3.3, ONNX |
|
|
| ONNX export of [BUT-FIT/DiCoW_v3_3](https://huggingface.co/BUT-FIT/DiCoW_v3_3) (revision `c34b64d9a9c5148c65fd355bb188d60343a6b44f`) for |
| [bolo](https://github.com/cb-Sigma/bolo), which runs it with ONNX Runtime and NumPy, without PyTorch. |
| DiCoW (Diarization-Conditioned Whisper) is Whisper large-v3-turbo conditioned on per-frame speaker activity |
| (silence / target / non-target / overlap masks), so a recording is transcribed once per diarized speaker, |
| overlapping speech included. The model is by the Speech@FIT group of Brno University of Technology and is |
| licensed CC-BY-4.0; this conversion keeps that licence. Changes: the weights are exported unchanged to two |
| graphs (encoder with FDDT conditioning and the decoder's cross-attention projections; decoder with a key/value |
| cache); the `*.int8.onnx` graphs are dynamically quantised (int8 weights, per output channel; the encoder's |
| feed-forward output matrices stay fp32). The CTC head is not included (the published generation settings do not |
| use it). On 24 AMI excerpts the int8 graphs scored the same cpWER as the PyTorch model (26.0 %); details in bolo's |
| docs/models.md, "DiCoW without PyTorch". |
|
|
| Exported with `server/tools/export_dicow_onnx.py` from the bolo repository; checksums in `dicow_onnx.json`. |
|
|
| If you use the model, please cite the papers its model card lists: |
|
|
| - A. Polok et al., "DiCoW: Diarization-conditioned Whisper for target speaker automatic speech recognition", |
| Computer Speech & Language 95, 2026, doi:10.1016/j.csl.2025.101841. |
| - A. Polok et al., "Target Speaker ASR with Whisper", ICASSP 2025, doi:10.1109/ICASSP49660.2025.10887683. |
| - A. Polok, D. Klement, S. Cornell, M. Wiesner, J. Černocký, S. Khudanpur, L. Burget, "SE-DiCoW: Self-Enrolled |
| Diarization-Conditioned Whisper", arXiv:2601.19194, 2026. |
|
|
