|
--- |
|
language: |
|
- ta |
|
metrics: |
|
- wer |
|
library_name: transformers |
|
pipeline_tag: automatic-speech-recognition |
|
--- |
|
|
|
# Model Card for Model ID |
|
|
|
<!-- Provide a quick summary of what the model is/does. --> |
|
|
|
This model is the fine-tuned version of Amrrs/wav2vec2-large-xlsr-53-tamil model for Tamil speech data. |
|
|
|
## Model Details |
|
|
|
|
|
## Uses |
|
|
|
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> |
|
This model can be used to fine-tune the wav2vec2-large model for Tamil language. |
|
|
|
#### Preprocessing [optional] |
|
|
|
* Resampling to 16000 Hz |
|
* Log-Mel Spectrogram extraction |
|
|
|
|
|
#### Training Hyperparameters |
|
|
|
training_args = TrainingArguments( |
|
output_dir="armms-xlsr", |
|
group_by_length=True, |
|
per_device_train_batch_size=4, |
|
gradient_accumulation_steps=3, |
|
evaluation_strategy="steps", |
|
num_train_epochs=30, |
|
gradient_checkpointing=True, |
|
fp16=True, |
|
seed = 42, |
|
save_steps=200, |
|
eval_steps=200, |
|
logging_steps=200, |
|
learning_rate=3e-4, |
|
warmup_steps=500, |
|
save_total_limit=2, |
|
push_to_hub=True, |
|
) |
|
|
