Nexdata's picture
Update README.md
2adbdaa verified
metadata
task_categories:
  - text-to-speech
language:
  - zh

Dataset Card for Nexdata/Chinese_Mandarin_Multi-emotional_Synthesis_Corpus

Description

22 People - Chinese Mandarin Multi-emotional Synthesis Corpus. It is recorded by Chinese native speaker, covering different ages and genders. six emotional text, and the syllables, phonemes and tones are balanced. Professional phonetician participates in the annotation. It precisely matches with the research and development needs of the speech synthesis.

For more details, please refer to the link: https://www.nexdata.ai/datasets/tts/1214?source=Huggingface

Specifications

Format

48,000Hz, 24bit, uncompressed wav, mono channel

Recording environment

professional recording studio

Recording content

seven emotions (happiness, anger, sadness, surprise, fear, disgust)

Speaker

22 persons, different age groups and genders

Device

microphone

Language

Mandarin

Annotation

word and pinyin transcription, prosodic boundary annotation

Application scenarios

speech synthesis

The amount of data

The amount of data for per person is 140 minutes, each emotion is 20 minutes

Licensing Information

Commercial License