Mandarin Chinese Seperated Track Spontaneous Dialogue Paralanguage Annotated Speech Synthesis Corpus, with a free dialogue style. Given a topic, the speaker can express themselves, and in each conversation, each person's audio is stored in their own separate WAV file. Professional linguists have annotated 16 types of paralanguage annotations, text annotations, timestamps, and other information to accurately match the research and development needs of speech synthesis.
For more details, please refer to the link: https://www.nexdata.ai/datasets/tts/1620?source=Github
48kHz, 24 bit, wav, mono channel
Recording studio
Spontaneous dialogue in given topics
294 people (Non-Professional Voice Actors) in total, gender balanced (144 females and 150 males), 18~60 years old
16 kinds of paralanguage annotation; text transcription; speaker ID, special symbol
Microphone
Mandarin Chinese
China(CHN)
zh-CN
Character Accuracy Rate 99%
Commercial License