Downloads · 30 days
16
5% of all-time downloads
wav2vec2-2-bart-large-no-adapter
sanchit-gandhi/wav2vec2-2-bart-large-no-adapter
wav2vec2-2-bart-large-no-adapter is a automatic speech recognition model from sanchit-gandhi. Use it when you need speech turned into text. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
332
Public
Repo size
72.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin2.3 GB · 90%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- Model type
- speech-encoder-decoder
- Access
- Public
- Created
- Mar 14, 2022
- Updated
- Mar 14, 2022
- SHA
- f09c487b
Datasets
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
This model was trained from scratch on the librispeech_asr dataset. It achieves the following results on the evaluation set:
- Loss: 5.6120
- Wer: 1.0267
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 5.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 6.7189 | 0.56 | 500 | 6.9796 | 0.9350 |
| 6.5068 | 1.12 | 1000 | 6.4823 | 1.3923 |
| 6.4601 | 1.68 | 1500 | 6.1801 | 1.1578 |
| 6.1802 | 2.24 | 2000 | 6.0002 | 1.7750 |
| 6.0888 | 2.8 | 2500 | 5.8453 | 1.7581 |
| 6.0993 | 3.36 | 3000 | 5.7702 | 1.4096 |
| 6.0851 | 3.92 | 3500 | 5.6634 | 1.0944 |
| 5.9357 | 4.48 | 4000 | 5.6120 | 1.0267 |
Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.2+cu113
- Datasets 1.18.3
- Tokenizers 0.11.0
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- speech-encoder-decoder
- Created
- Mar 14, 2022
- Updated
- Mar 14, 2022