Downloads · 30 days
12
13% of all-time downloads
wav2vec2-large-xls-r-300m-bengali-v8
rashedsafa/wav2vec2-large-xls-r-300m-bengali-v8
wav2vec2-large-xls-r-300m-bengali-v8 is a automatic speech recognition model from rashedsafa. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
94
Public
Repo size
10.1 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin1.3 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Aug 22, 2022
- Updated
- Aug 23, 2022
- SHA
- 543b35c5
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. -->
wav2vec2-large-xls-r-300m-bengali-v8
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice dataset. It achieves the following results on the evaluation set:
- Loss: 0.7874
- Wer: 0.6777
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: 9e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- 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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 7.2332 | 0.85 | 400 | 3.3381 | 1.0 |
| 2.3574 | 1.71 | 800 | 0.8236 | 0.7516 |
| 0.8096 | 2.56 | 1200 | 0.9337 | 0.6717 |
| 1.1487 | 3.41 | 1600 | 1.1691 | 0.7665 |
| 0.806 | 4.26 | 2000 | 0.7716 | 0.6642 |
| 0.7746 | 5.12 | 2400 | 0.7874 | 0.6777 |
| 0.7736 | 5.97 | 2800 | 0.7874 | 0.6777 |
| 0.775 | 6.82 | 3200 | 0.7874 | 0.6777 |
| 0.7718 | 7.68 | 3600 | 0.7874 | 0.6777 |
| 0.7757 | 8.53 | 4000 | 0.7874 | 0.6777 |
| 0.7761 | 9.38 | 4400 | 0.7874 | 0.6777 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Aug 22, 2022
- Updated
- Aug 23, 2022