README.md

metadata

pretty_name: Representative Samples
language:
  - vi
license: cc-by-4.0
task_categories:
  - audio-classification
tags:
  - audio
  - speech
  - speaker-recognition
  - speaker-verification
  - vietnamese
configs:
  - config_name: default
    data_files:
      - split: train
        path: train/**/*
      - split: validation
        path: val/**/*

Representative Samples

This repository contains representative Vietnamese speech samples organized by speaker folders.

Dataset structure

Representative_Samples/

• train/
  • spk1/
    • *.wav
  • spk2/
    • *.wav
• val/
  • spk1/
    • *.wav
  • spk2/
    • *.wav

Fields

• audio: audio waveform file.
• label: speaker folder name, for example spk1, spk2, etc.

Splits

• train: training representative samples.
• validation: validation representative samples.

Intended use

Representative Vietnamese speech samples for speech and speaker-related research.