Automatic Speech Recognition
speechbrain
PyTorch
English
wav2vec2
CTC
Attention
Transformer
hf-asr-leaderboard
Eval Results (legacy)
Eval Results
Instructions to use speechbrain/asr-wav2vec2-librispeech with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- speechbrain
How to use speechbrain/asr-wav2vec2-librispeech with speechbrain:
from speechbrain.pretrained import EncoderASR model = EncoderASR.from_hparams( "speechbrain/asr-wav2vec2-librispeech" ) model.transcribe_file("file.wav") - Notebooks
- Google Colab
- Kaggle
Update hyperparams_develop.yaml
Browse files- hyperparams_develop.yaml +2 -1
hyperparams_develop.yaml
CHANGED
|
@@ -54,7 +54,8 @@ encoder: !new:speechbrain.nnet.containers.LengthsCapableSequential
|
|
| 54 |
enc: !ref <enc>
|
| 55 |
ctc_lin: !ref <ctc_lin>
|
| 56 |
|
| 57 |
-
|
|
|
|
| 58 |
|
| 59 |
modules:
|
| 60 |
encoder: !ref <encoder>
|
|
|
|
| 54 |
enc: !ref <enc>
|
| 55 |
ctc_lin: !ref <ctc_lin>
|
| 56 |
|
| 57 |
+
decoding_function: !name:speechbrain.decoders.ctc_greedy_decode
|
| 58 |
+
blank_id: !ref <blank_index>
|
| 59 |
|
| 60 |
modules:
|
| 61 |
encoder: !ref <encoder>
|