---
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
datasets:
- lj_speech
model-index:
- name: speecht5_finetuned_English
  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_English

This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the lj_speech dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3717

## 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: 0.0001
- train_batch_size: 4
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- 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: 100
- training_steps: 1500
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 3.772         | 0.3053 | 100  | 0.4201          |
| 3.5912        | 0.6107 | 200  | 0.4052          |
| 3.4604        | 0.9160 | 300  | 0.3948          |
| 3.3894        | 1.2214 | 400  | 0.3906          |
| 3.3737        | 1.5267 | 500  | 0.3865          |
| 3.3628        | 1.8321 | 600  | 0.3851          |
| 3.3236        | 2.1374 | 700  | 0.3821          |
| 3.306         | 2.4427 | 800  | 0.3811          |
| 3.2859        | 2.7481 | 900  | 0.3797          |
| 3.2663        | 3.0534 | 1000 | 0.3763          |
| 3.2368        | 3.3588 | 1100 | 0.3757          |
| 3.2107        | 3.6641 | 1200 | 0.3749          |
| 3.2035        | 3.9695 | 1300 | 0.3730          |
| 3.1969        | 4.2748 | 1400 | 0.3728          |
| 3.2107        | 4.5802 | 1500 | 0.3717          |


### Framework versions

- Transformers 4.46.0.dev0
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.20.1