Instructions to use espnet/owsm_v4_base_102M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ESPnet
How to use espnet/owsm_v4_base_102M with ESPnet:
import soundfile from espnet2.bin.asr_inference import Speech2Text model = Speech2Text.from_pretrained( "espnet/owsm_v4_base_102M" ) speech, rate = soundfile.read("speech.wav") text, *_ = model(speech)[0] - Notebooks
- Google Colab
- Kaggle
Commit History
13 days ago
-
Link the community audio.cpp GGUF builds
Verified signatureAdds a pointer from the model card to the community GGUF builds at audio-cpp/audio.cpp-gguf, so people who want to run OWSM on CPU without Python find them from the official page. The builds are produced and maintained by the audio.cpp project, not by ESPnet, and the added sentence says so. Nothing is hosted here.55c92be
-
Link a demo that runs, and one for this version
Verified signatureThe card pointed at pyf98/OWSM_v3_demo, which has been erroring since 2024 and demonstrates v3 on a v4 model card. espnet/owsm-v4 is maintained from egs2/owsm_v4/s2t1/demo in the espnet repository; it serves the 1B checkpoint rather than this one, so the line says so. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>1529407
-
Add a usage example to the model card
Verified signature96dbb52
-
Update README.md
Verified signature1652139 -
Update README.md
Verified signature0157a38
-
Add library name and pipeline tag (#1)
Verified signature- Add library name and pipeline tag (43b0f52fd794c3183320c21fb4f1bf4aad97d7b1) Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>a767ef3
-
Update README.md
Verified signature4290277 -
Update README.md
Verified signature5128950
-
Update README.md
Verified signature6e14d40
-
add logs
5594659 -
Update README.md
Verified signature6f42746
-
Update README.md
Verified signature4798804 -
Update README.md
Verified signaturec14c548 -
add model
084ce5f -
initial commit
Verified signature42b88a4