File size: 1,709 Bytes
5c3592c d1b7543 5c3592c d1b7543 5c3592c d1b7543 5c3592c d1b7543 5c3592c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 |
---
language:
- ta
license: apache-2.0
tags:
- automatic-speech-recognition
- common_voice
- generated_from_trainer
datasets:
- common_voice
model-index:
- name: wav2vec2-common_voice-tamil
results: []
---
<!-- 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-common_voice-tamil
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the COMMON_VOICE - TA dataset.
It achieves the following results on the evaluation set:
- Loss: 1.1172
- Wer: 1.0070
## 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: 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: 3.0
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| No log | 0.84 | 100 | 4.0148 | 1.0 |
| No log | 1.69 | 200 | 3.1738 | 1.0 |
| No log | 2.54 | 300 | 2.5980 | 1.0236 |
### Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.1+cu113
- Datasets 1.18.1.dev0
- Tokenizers 0.10.3
|