metadata
language:
- eng
license: apache-2.0
tags:
- '[finetuned_model, lj_speech11]'
- generated_from_trainer
base_model: facebook/wav2vec2-base-960h
datasets:
- FYP/LJ-SpeechLJ
model-index:
- name: SpeechT5 STT Wav2Vec2
results: []
SpeechT5 STT Wav2Vec2
This model is a fine-tuned version of facebook/wav2vec2-base-960h on the Lj-Speech dataset. It achieves the following results on the evaluation set:
- Loss: 310.7488
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- 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
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3527.1737 | 0.5362 | 50 | 433.5859 |
| 441.3472 | 1.0724 | 100 | 331.3980 |
| 404.9114 | 1.6086 | 150 | 379.2116 |
| 330.0151 | 2.1448 | 200 | 319.7489 |
| 346.7672 | 2.6810 | 250 | 307.3814 |
| 398.5639 | 3.2172 | 300 | 349.2720 |
| 331.2473 | 3.7534 | 350 | 308.0336 |
| 311.4908 | 4.2895 | 400 | 306.9877 |
| 312.0619 | 4.8257 | 450 | 310.7488 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1