metadata
library_name: transformers
license: mit
base_model: facebook/w2v-bert-2.0
tags:
- automatic-speech-recognition
- BembaSpeech
- generated_from_trainer
metrics:
- wer
model-index:
- name: w2v-bert-bem-bl
results: []
w2v-bert-bem-bl
This model is a fine-tuned version of facebook/w2v-bert-2.0 on the BEMBASPEECH - BEM dataset. It achieves the following results on the evaluation set:
- Loss: 0.2136
- Wer: 0.4539
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.5344 | 0.7027 | 500 | 0.5448 | 0.7379 |
| 0.4242 | 1.4055 | 1000 | 0.3055 | 0.6025 |
| 0.3603 | 2.1082 | 1500 | 0.2693 | 0.5385 |
| 0.3144 | 2.8110 | 2000 | 0.2683 | 0.5529 |
| 0.2656 | 3.5137 | 2500 | 0.2472 | 0.5258 |
| 0.2311 | 4.2164 | 3000 | 0.2352 | 0.5026 |
| 0.2106 | 4.9192 | 3500 | 0.2327 | 0.5003 |
| 0.1816 | 5.6219 | 4000 | 0.2298 | 0.4987 |
| 0.1432 | 6.3247 | 4500 | 0.2178 | 0.4686 |
| 0.1431 | 7.0274 | 5000 | 0.2172 | 0.4747 |
| 0.1069 | 7.7301 | 5500 | 0.2136 | 0.4539 |
| 0.0767 | 8.4329 | 6000 | 0.2270 | 0.4403 |
| 0.0667 | 9.1356 | 6500 | 0.2375 | 0.4385 |
| 0.0468 | 9.8384 | 7000 | 0.2403 | 0.4353 |
Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1