Automatic Speech Recognition
Transformers
PyTorch
TensorFlow
JAX
Safetensors
English
whisper
audio
hf-asr-leaderboard
Eval Results (legacy)
Eval Results
Instructions to use openai/whisper-medium.en with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use openai/whisper-medium.en with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="openai/whisper-medium.en")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("openai/whisper-medium.en") model = AutoModelForSpeechSeq2Seq.from_pretrained("openai/whisper-medium.en", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Correct long-form generation config parameters 'max_initial_timestamp_index' and 'prev_sot_token_id'. (#16)
Verified signature- Correct long-form generation config parameters 'max_initial_timestamp_index' and 'prev_sot_token_id'. (6b72fc8b2ebbc86d6a75d8520ee48ebb0f1f5c32)2e98eb6
-
Adding `safetensors` variant of this model (#11)
- Adding `safetensors` variant of this model (363c110a7fe5682b70b65adf501045d72961b815) Co-authored-by: Nicolas Patry <Narsil@users.noreply.huggingface.co>b108f11
-
Update generation config with word-level alignment heads (#10)
- Upload config (193b730abf32bbf3c1defb98f4dff835aed61e56)32d2c4d
-
Update README.md
55e79ed
-
Update generation_config.json to suppress task tokens (#8)
- Update generation_config.json to suppress task tokens (e4a3747abff407353b402d1202459a0b49474068) Co-authored-by: Guillaume Klein <guillaumekln@users.noreply.huggingface.co>960623b -
Update config.json to suppress task tokens (#7)
- Update config.json to suppress task tokens (01a0a57e9ea7b0685cc43df612ba1a68478f6c17) Co-authored-by: Guillaume Klein <guillaumekln@users.noreply.huggingface.co>c94c8bb
-
Add Flax weights
sanchit-gandhid304ad8
-
Update README.md
29963c9
-
Update README.md (#6)
- Update README.md (a43c9dc3eafffa30c6ad3bea53620565a79f9c7c)a8f3af2
-
Changes for fast tokenizer (#5)
- Add tokenizer.json (9d44cb6bf8fbd16ae5d2c0191debf5717fc5f466) Co-authored-by: Jonatan Kłosko <jonatanklosko@users.noreply.huggingface.co>e3bd65e
-
Update generation_config.json
1819ebb -
Upload config
a5c7144
-
Revert: Add "<|startoftranscript|>" to forced decoder ids (#4)
- Revert: Add "<|startoftranscript|>" to forced decoder ids (7b96b50affcb13d65a9e7a2bbd37edd580a2c2d7)5ae922f -
Add "<|startoftranscript|>" to forced decoder ids (#3)
- Add "<|startoftranscript|>" to forced decoder ids (9a3079ffc21d00e3012d77a1a40a4ce603838025)58d9062
-
Allow single quotes "'" and hyphens "-" (#2)
- Allow single quotes "'" and hyphens "-" (523f7db29d28c53cf8446d008efbaea2595baa1c) Co-authored-by: Sanchit Gandhi <sanchit-gandhi@users.noreply.huggingface.co>f0a02de
-
Convert WERs from decimal to percentage
e5ea493
-
Create README.md
929e4fd -
Update `max_length` param
2fe5eef -
Update `return_mask` param
33335f2
-
Add TF weights (#1)
- Add TF weights (1b47e5177eb459fc5de32866d2222bf4dbe8f3f9)5be1727
-
Upload config
ef162df -
Upload tokenizer
a214e82 -
Upload config
fc3fb16 -
Upload tokenizer
ef34083 -
Upload tokenizer
9261f7d -
Upload tokenizer
78cba71 -
Upload tokenizer
b6f6f16
-
Upload processor
32e1749 -
Upload config
a5d54a5 -
Upload processor
3d15904 -
Upload config
73142b4 -
Upload processor
6f3a63e -
Upload processor
f6bf9d1 -
Upload config
85cf917
-
Upload tokenizer
717caee -
Upload WhisperForConditionalGeneration
31a4aab -
Upload processor
7368b8e -
Upload config
0fde80b
-
Upload processor
13d8f54 -
Upload model
c6da634 -
initial commit
1b46166