---
library_name: transformers
language:
- sw
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_17_0
model-index:
- name: Swahili TTS
  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. -->

# Swahili TTS

This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the Swahili dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5318

## 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 OptimizerNames.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 |
|:-------------:|:------:|:----:|:---------------:|
| 0.6394        | 1.3351 | 1000 | 0.5669          |
| 0.591         | 2.6702 | 2000 | 0.5440          |
| 0.5796        | 4.0053 | 3000 | 0.5352          |
| 0.5689        | 5.3405 | 4000 | 0.5318          |


### Framework versions

- Transformers 4.50.0.dev0
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0