metadata
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: English_Technical_finetuned
results: []
datasets:
- Tejasva-Maurya/English-Technical-Speech-Dataset
language:
- en
English_Technical_finetuned
This model is a fine-tuned version of microsoft/speecht5_tts on an English Technical Speech Dataset . It achieves the following results on the evaluation set:
- Loss: 0.4451
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6122 | 0.3168 | 100 | 0.5289 |
| 0.5468 | 0.6337 | 200 | 0.4885 |
| 0.5207 | 0.9505 | 300 | 0.4745 |
| 0.5086 | 1.2673 | 400 | 0.4729 |
| 0.5012 | 1.5842 | 500 | 0.4638 |
| 0.4982 | 1.9010 | 600 | 0.4564 |
| 0.4888 | 2.2178 | 700 | 0.4528 |
| 0.4862 | 2.5347 | 800 | 0.4515 |
| 0.4866 | 2.8515 | 900 | 0.4454 |
| 0.4753 | 3.1683 | 1000 | 0.4451 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Tokenizers 0.19.1