---
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: speecht5_finetuned_tts_tr
  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. -->

# speecht5_finetuned_tts_tr

This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5704

## 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: 0.0005
- train_batch_size: 4
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 8
- 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: cosine
- lr_scheduler_warmup_steps: 150
- training_steps: 800
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.7671        | 1.1629 | 200  | 0.7422          |
| 0.6787        | 2.3258 | 400  | 0.6538          |
| 0.6045        | 3.4887 | 600  | 0.6151          |
| 0.5662        | 4.6516 | 800  | 0.5704          |


### Framework versions

- Transformers 4.55.2
- Pytorch 2.6.0+cu124
- Datasets 4.0.0
- Tokenizers 0.21.4