ShawnPi233/SynParaSpeech

Official Repository of Paper: "SynParaSpeech: Automated Synthesis of Paralinguistic Datasets for Speech Generation and Understanding" (ICASSP 2026)

33
/ 100
Emerging

This project helps researchers and developers create realistic speech generation and understanding systems that include natural human sounds like laughter, sighs, and gasps. It automatically produces high-quality datasets consisting of audio clips, corresponding text, and precise timing for these 'paralinguistic' sounds. Speech AI developers, voice model trainers, and academic researchers focused on natural language processing would use this.

Use this if you need large-scale, accurately annotated datasets containing paralinguistic sounds to train or evaluate speech synthesis (Text-to-Speech) or speech understanding models.

Not ideal if you need a dataset for general speech recognition without specific emphasis on paralinguistic events or if your primary language of interest is not Chinese.

speech-synthesis speech-recognition natural-language-processing voice-AI machine-learning-datasets
No License No Package No Dependents
Maintenance 10 / 25
Adoption 8 / 25
Maturity 7 / 25
Community 8 / 25

How are scores calculated?

Stars

66

Forks

4

Language

JavaScript

License

Last pushed

Jan 27, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/ShawnPi233/SynParaSpeech"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.