| anis | Aniskhan29 | draft | 0 | disc |
| Faster + more accurate alternative: FunASR SenseVoice (234MB, 170x realtime) | langgz | open | 0 | disc |
| Model card update suggested | cfasana | open | 0 | disc |
| WER is large on test other? | AICoding91 | open | 0 | disc |
| deccan2k | deccan2k | draft | 0 | disc |
| Why does snapshot_download for whisper-small download several GB when the PyTorch model is only ~500 MB? | Kerwin11 | open | 0 | disc |
| Upload 336.mp4 | anor337 | open | 0 | disc |
| TFLite Model Translates Non-English Transcription Automatically | rabyunghwa | open | 0 | disc |
| Long-Form Translation!! | scj0709 | open | 0 | disc |
| Upload model.safetensors | eknathtecho | open | 0 | disc |
| How is whisper-small larger than whisper-base? 967 MB vs 290 MB | apssg96 | open | 0 | disc |
| [AUTOMATED] Model Memory Requirements | model-sizer-bot | open | 0 | disc |
| [AUTOMATED] Model Memory Requirements | model-sizer-bot | open | 0 | disc |
| Add missing merge to tokenizer | sanchit-gandhi | merged | 0 | disc |
| Correct long-form generation config parameters 'max_initial_timestamp_index' and 'prev_sot_token_id'. | patrickvonplaten | merged | 0 | disc |
| Whisper processor decoder boosted with N-grams | kamil | open | 0 | disc |
| Correct added token ids | sanchit-gandhi | open | 0 | disc |
| update config of preprocessor | Pranjal12345 | open | 0 | disc |
| add special tokens for fast | sanchit-gandhi | merged | 0 | disc |
| add timestamp tokens | sanchit-gandhi | closed | 0 | disc |
| add timestamp tokens | sanchit-gandhi | merged | 0 | disc |
| Adding `safetensors` variant of this model | Narsil | merged | 0 | disc |
| Update generation config with word-level alignment heads | sanchit-gandhi | merged | 0 | disc |
| Adding `safetensors` variant of this model | Narsil | merged | 0 | disc |
| Clarifying bos_token | taohoang | open | 0 | disc |
| Update README.md | deleted | closed | 0 | disc |
| Unable to load saved model from "model-directory" using pipeline | MLLife | closed | 0 | disc |
| Sagemaker Payload limit issue (413) | MLLife | open | 0 | disc |
| Update generation_config.json to suppress task tokens | guillaumekln | merged | 0 | disc |
| Update config.json to suppress task tokens | guillaumekln | open | 0 | disc |
| suppress-task-tokens | guillaumekln | closed | 0 | disc |
| Adding ONNX file of this model | lizx | open | 0 | disc |
| Update the pad token | Narsil | merged | 0 | disc |
| Google Coral | nastasev | open | 0 | disc |
| Upload tokenizer.json | ArthurZ | merged | 0 | disc |
| Upload tokenizer | ArthurZ | closed | 0 | disc |
| Forcing decoder to transcribe English is not working | lucas-aixplain | open | 0 | disc |
| Update README.md | sanchit-gandhi | merged | 0 | disc |
| change hebrew code from iw (deprecated) to he | altryne | merged | 0 | disc |
| change hebrew code from iw (deprecated) to he | altryne | closed | 0 | disc |
| Changes for fast tokenizer | jonatanklosko | merged | 0 | disc |
| Upload config | ArthurZ | merged | 0 | disc |
| Mismatched outputs from encoders of `transformers` and `whisper` | JinchaoLove | open | 0 | disc |
| Fix unclosed parenthesis on logits line of example code on the English to english section | Camisa | closed | 0 | disc |
| Closed bracket in whisper-small | cupofsanity | open | 0 | disc |
| Revert forced decoder ids | sanchit-gandhi | merged | 0 | disc |
| Add "<|startoftranscript|>" to forced decoder ids | sanchit-gandhi | merged | 0 | disc |
| Allow single quotes "'" and hyphens "-" | sanchit-gandhi | merged | 0 | disc |
| Pred and ref are switched in the WER calculation. | lichenda | open | 0 | disc |
| How to set the language in the Inference API headers? | chameleon-lizard | open | 0 | disc |
| Add TF weights | amyeroberts | merged | 0 | disc |