Downloads · 30 days
4
14% of all-time downloads
wav2vec2-balti
YuvrajGujari/wav2vec2-balti
wav2vec2-balti is a automatic speech recognition model from YuvrajGujari. 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
29
Public
Parameters
316M
1.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.3 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Aug 7, 2026
- Updated
- Aug 7, 2026
- SHA
- 1efd0235
Base models
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-balti
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.3484
- Wer: 0.2282
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 300
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 24.9571 | 0.3971 | 200 | 3.1221 | 1.0 |
| 12.3515 | 0.7942 | 400 | 1.1488 | 0.8080 |
| 7.2630 | 1.1906 | 600 | 0.6948 | 0.6025 |
| 5.8629 | 1.5877 | 800 | 0.5714 | 0.5066 |
| 5.5451 | 1.9849 | 1000 | 0.4945 | 0.4236 |
| 4.7129 | 2.3812 | 1200 | 0.4673 | 0.4094 |
| 4.6447 | 2.7784 | 1400 | 0.4284 | 0.3989 |
| 3.8591 | 3.1747 | 1600 | 0.4073 | 0.3692 |
| 3.8259 | 3.5719 | 1800 | 0.3952 | 0.3406 |
| 3.6229 | 3.9690 | 2000 | 0.3744 | 0.3340 |
| 3.1691 | 4.3654 | 2200 | 0.3719 | 0.3179 |
| 3.2988 | 4.7625 | 2400 | 0.3513 | 0.3078 |
| 2.7646 | 5.1588 | 2600 | 0.3599 | 0.3105 |
| 2.7120 | 5.5560 | 2800 | 0.3445 | 0.2917 |
| 2.8528 | 5.9531 | 3000 | 0.3358 | 0.2842 |
| 2.3170 | 6.3495 | 3200 | 0.3455 | 0.2882 |
| 2.6743 | 6.7466 | 3400 | 0.3492 | 0.2862 |
| 2.1990 | 7.1430 | 3600 | 0.3379 | 0.2769 |
| 2.2063 | 7.5401 | 3800 | 0.3429 | 0.2731 |
| 2.3009 | 7.9372 | 4000 | 0.3361 | 0.2734 |
| 1.9231 | 8.3336 | 4200 | 0.3332 | 0.2639 |
| 1.8913 | 8.7307 | 4400 | 0.3247 | 0.2621 |
| 1.6484 | 9.1271 | 4600 | 0.3292 | 0.2489 |
| 1.7733 | 9.5242 | 4800 | 0.3208 | 0.2592 |
| 1.6847 | 9.9213 | 5000 | 0.3327 | 0.2495 |
| 1.6250 | 10.3177 | 5200 | 0.3416 | 0.2458 |
| 1.7336 | 10.7148 | 5400 | 0.3212 | 0.2456 |
| 1.3292 | 11.1112 | 5600 | 0.3407 | 0.2393 |
| 1.4187 | 11.5083 | 5800 | 0.3319 | 0.2386 |
| 1.3104 | 11.9054 | 6000 | 0.3266 | 0.2378 |
| 1.2938 | 12.3018 | 6200 | 0.3373 | 0.2372 |
| 1.0719 | 12.6989 | 6400 | 0.3469 | 0.2326 |
| 1.1590 | 13.0953 | 6600 | 0.3380 | 0.2297 |
| 1.1588 | 13.4924 | 6800 | 0.3409 | 0.2323 |
| 1.0949 | 13.8896 | 7000 | 0.3386 | 0.2289 |
| 1.1241 | 14.2859 | 7200 | 0.3354 | 0.2284 |
| 1.0928 | 14.6830 | 7400 | 0.3484 | 0.2282 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Aug 7, 2026
- Updated
- Aug 7, 2026