Automatic Speech Recognition
Transformers
PyTorch
TensorFlow
JAX
English
wav2vec2
speech
audio
hf-asr-leaderboard
Eval Results (legacy)
Eval Results
Instructions to use facebook/wav2vec2-large-960h-lv60-self with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/wav2vec2-large-960h-lv60-self with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-large-960h-lv60-self")# Load model directly from transformers import AutoProcessor, AutoModelForCTC processor = AutoProcessor.from_pretrained("facebook/wav2vec2-large-960h-lv60-self") model = AutoModelForCTC.from_pretrained("facebook/wav2vec2-large-960h-lv60-self") - Notebooks
- Google Colab
- Kaggle
Update results
Browse files
README.md
CHANGED
|
@@ -15,13 +15,30 @@ model-index:
|
|
| 15 |
name: Automatic Speech Recognition
|
| 16 |
type: automatic-speech-recognition
|
| 17 |
dataset:
|
| 18 |
-
name:
|
| 19 |
type: librispeech_asr
|
| 20 |
-
|
|
|
|
|
|
|
|
|
|
| 21 |
metrics:
|
| 22 |
- name: Test WER
|
| 23 |
type: wer
|
| 24 |
value: 1.9
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
---
|
| 26 |
|
| 27 |
# Wav2Vec2-Large-960h-Lv60 + Self-Training
|
|
|
|
| 15 |
name: Automatic Speech Recognition
|
| 16 |
type: automatic-speech-recognition
|
| 17 |
dataset:
|
| 18 |
+
name: LibriSpeech (clean)
|
| 19 |
type: librispeech_asr
|
| 20 |
+
config: clean
|
| 21 |
+
split: test
|
| 22 |
+
args:
|
| 23 |
+
language: en
|
| 24 |
metrics:
|
| 25 |
- name: Test WER
|
| 26 |
type: wer
|
| 27 |
value: 1.9
|
| 28 |
+
- task:
|
| 29 |
+
name: Automatic Speech Recognition
|
| 30 |
+
type: automatic-speech-recognition
|
| 31 |
+
dataset:
|
| 32 |
+
name: LibriSpeech (other)
|
| 33 |
+
type: librispeech_asr
|
| 34 |
+
config: other
|
| 35 |
+
split: test
|
| 36 |
+
args:
|
| 37 |
+
language: en
|
| 38 |
+
metrics:
|
| 39 |
+
- name: Test WER
|
| 40 |
+
type: wer
|
| 41 |
+
value: 3.9
|
| 42 |
---
|
| 43 |
|
| 44 |
# Wav2Vec2-Large-960h-Lv60 + Self-Training
|