README.md

metadata

license: cc-by-nc-4.0
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: wav2vec2-base-finetune-vi-v2
    results: []
widget:
  - example_title: SOICT 2023 - SLU public test 1
    src: >-
      https://huggingface.co/foxxy-hm/wav2vec2-base-finetune-vi/raw/main/audio-test/055R7BruAa333g9teFfamQH.wav
  - example_title: SOICT 2023 - SLU public test 2
    src: >-
      https://huggingface.co/foxxy-hm/wav2vec2-base-finetune-vi/raw/main/audio-test/0BLHhoJexE8THB8BrsZxWbh.wav
  - example_title: SOICT 2023 - SLU public test 3
    src: >-
      https://huggingface.co/foxxy-hm/wav2vec2-base-finetune-vi/raw/main/audio-test/1ArUTGWJQ9YALH2xaNhU6GV.wav

wav2vec2-base-finetune-vi-v2

This model is a fine-tuned version of nguyenvulebinh/wav2vec2-base-vietnamese-250h on the None dataset. It achieves the following results on the evaluation set:

• Loss: 4.6968
• Wer: 0.9698

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
17.4813	0.67	500	5.4377	0.9996
5.4042	1.34	1000	5.0291	0.9996
4.836	2.0	1500	4.9688	0.9694
4.7651	2.67	2000	4.6968	0.9698

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3