|
|
--- |
|
|
library_name: transformers |
|
|
license: mit |
|
|
base_model: microsoft/speecht5_tts |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: speecht5_finetuned_interview |
|
|
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_emirhan_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: 0.4392 |
|
|
|
|
|
## Model description |
|
|
|
|
|
The Speech T5 model is a text-to-speech (TTS) model based on the T5 architecture. It has been |
|
|
pretrained on a large corpus of speech data, allowing it to understand and generate human-like |
|
|
speech from input text. The model is capable of handling various speech synthesis tasks, making it |
|
|
suitable for applications such as virtual assistants, audiobook production, and more |
|
|
|
|
|
|
|
|
## Intended uses & limitations |
|
|
|
|
|
More information needed |
|
|
|
|
|
## Training and evaluation data |
|
|
|
|
|
The model was trained using a custom-made dataset of 170 audio samples, containing commonly asked interview lines. Synthetic audio was generated using Amazon AWS Polly, which offered diverse voice options. The dataset was carefully curated to ensure a variety of speech styles, accents, and phonetic structures, enhancing the model's ability to generalize. |
|
|
|
|
|
|
|
|
|
|
|
### Training hyperparameters |
|
|
|
|
|
The following hyperparameters were used during training: |
|
|
- learning_rate: 2e-05 |
|
|
- train_batch_size: 2 |
|
|
- eval_batch_size: 1 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 4 |
|
|
- total_train_batch_size: 8 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_steps: 40 |
|
|
- training_steps: 250 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-------:|:----:|:---------------:| |
|
|
| 0.6078 | 2.2535 | 40 | 0.4783 | |
|
|
| 0.5393 | 4.5070 | 80 | 0.4533 | |
|
|
| 0.4864 | 6.7606 | 120 | 0.4480 | |
|
|
| 0.4846 | 9.0141 | 160 | 0.4493 | |
|
|
| 0.4628 | 11.2676 | 200 | 0.4383 | |
|
|
| 0.4731 | 13.5211 | 240 | 0.4392 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.45.0.dev0 |
|
|
- Pytorch 2.4.1+cu118 |
|
|
- Datasets 3.0.0 |
|
|
- Tokenizers 0.20.0 |
|
|
|
