The dataset is currently empty. Upload or create new data files. Then, you will be able to explore them in the Dataset Viewer.

Description

2 People - Chinese Natural Conversation Speech Synthesis Corpus. It is recorded by Chinese native speaker, natural conversation style. phonemes and tones are balanced. Professional phonetician participates in the annotation, and annotate secondary language, Secondary Language Annotation: Inhalation: V; Pause: P; Hesitation: T; Mouth clicking: M; Drawl: D; Cough: C; Laughter: L; Stutter repetition: R; Inversion: I; Modal particle: S (Modal particles include "ah", "oh", "wow", "right?", "what?", "well" etc.).

For more details, please refer to the link: https://www.nexdata.ai/datasets/tts/1470?source=huggingface

Specifications

Format

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

Recording environment

professional recording studio;

Recording content

Given topic, speaker's self-expression;

Speaker

professional Character Voice, natural conversation, not reading aloud, one male voice, one female voice;

Annotation

word and pinyin transcription, prosodic boundary annotation, secondary language annotation;

Device

microphone;

Language

chinese, chinese/english mixed;

Application scenarios

speech synthesis

Licensing Information

Commercial License

Downloads last month
6