Downloads · 30 days
6
12% of all-time downloads
IEEEAccess_Previous
kaarthu2003/IEEEAccess_Previous
IEEEAccess_Previous is a automatic speech recognition model from kaarthu2003. 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
49
Public
Parameters
316M
39.1 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
- May 16, 2025
- Updated
- May 17, 2025
- SHA
- 717af51e
Base models
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-IEEEAccess-FinalRun
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: inf
- Wer: 0.6281
- Cer: 0.2261
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: 32
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Use adamw_torch 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: 1200
- num_epochs: 45
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 3.3927 | 2.0101 | 1200 | inf | 0.7921 | 0.3049 |
| 0.8527 | 4.0201 | 2400 | inf | 0.7935 | 0.3785 |
| 0.6921 | 6.0302 | 3600 | inf | 0.6959 | 0.2665 |
| 0.5897 | 8.0402 | 4800 | inf | 0.6303 | 0.2240 |
| 0.5151 | 10.0503 | 6000 | inf | 0.6588 | 0.2535 |
| 0.4529 | 12.0604 | 7200 | inf | 0.6340 | 0.2450 |
| 0.4 | 14.0704 | 8400 | inf | 0.6539 | 0.2383 |
| 0.357 | 16.0805 | 9600 | inf | 0.6479 | 0.2327 |
| 0.3155 | 18.0905 | 10800 | inf | 0.6432 | 0.2248 |
| 0.277 | 20.1006 | 12000 | inf | 0.6441 | 0.2361 |
| 0.2475 | 22.1106 | 13200 | inf | 0.6541 | 0.2431 |
| 0.2249 | 24.1207 | 14400 | inf | 0.6653 | 0.2500 |
| 0.2011 | 26.1308 | 15600 | inf | 0.6421 | 0.2463 |
| 0.1802 | 28.1408 | 16800 | inf | 0.6387 | 0.2323 |
| 0.1649 | 30.1509 | 18000 | inf | 0.6155 | 0.2262 |
| 0.1501 | 32.1609 | 19200 | inf | 0.6381 | 0.2331 |
| 0.1356 | 34.1710 | 20400 | inf | 0.6422 | 0.2345 |
| 0.1224 | 36.1811 | 21600 | inf | 0.6306 | 0.2305 |
| 0.1103 | 38.1911 | 22800 | inf | 0.6339 | 0.2294 |
| 0.1014 | 40.2012 | 24000 | inf | 0.6311 | 0.2275 |
| 0.0934 | 42.2112 | 25200 | inf | 0.6258 | 0.2244 |
| 0.0875 | 44.2213 | 26400 | inf | 0.6281 | 0.2261 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.1.0+cu118
- Datasets 3.6.0
- Tokenizers 0.21.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- May 16, 2025
- Updated
- May 17, 2025