| --- |
| library_name: transformers |
| license: mit |
| base_model: microsoft/speecht5_tts |
| tags: |
| - generated_from_trainer |
| datasets: |
| - Chithekitale/virus_one_and_two |
| model-index: |
| - name: Chichewa TTS_final |
| 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. --> |
|
|
| # Chichewa TTS_final |
| |
| This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the chichewaspeechset dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.3945 |
| |
| ## 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: 1e-05 |
| - train_batch_size: 16 |
| - eval_batch_size: 8 |
| - seed: 42 |
| - gradient_accumulation_steps: 2 |
| - total_train_batch_size: 32 |
| - 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: 500 |
| - training_steps: 4000 |
| - mixed_precision_training: Native AMP |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | |
| |:-------------:|:-------:|:----:|:---------------:| |
| | 0.4413 | 6.9965 | 1000 | 0.4014 | |
| | 0.431 | 13.9895 | 2000 | 0.3980 | |
| | 0.4227 | 20.9825 | 3000 | 0.3953 | |
| | 0.4154 | 27.9754 | 4000 | 0.3945 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 4.50.3 |
| - Pytorch 2.6.0+cu118 |
| - Datasets 3.5.0 |
| - Tokenizers 0.21.1 |
| |
