1 dépôt
Using written scripts to improve the accuracy of phonetic analysis of spoken audio.
Distinct from Text Processing: Specifically uses text as a guide for audio recognition, not general text manipulation.
Explore 1 awesome GitHub repository matching part of an awesome list · Dialogue-Guided Recognition. Refine with filters or upvote what's useful.
Rhubarb is an automated lip sync generator and phonetic speech analyzer that converts audio recordings into timed mouth-shape animation data. It identifies sounds and syllables within audio files to map them to specific visual mouth shapes, serving as an animation timing exporter for external character animation software. The tool utilizes a language-independent phonetic recognizer to process speech regardless of the spoken language. To increase accuracy, it supports dialogue-guided recognition by using external text files to guide the phonetic analysis of specific spoken scripts. The system
Uses external text files to improve phonetic recognition and handle unknown word pronunciations during analysis.