search

X-LANCE / VoiceFlow-TTS

[ICASSP 2024] This is the official code for "VoiceFlow: Efficient Text-to-Speech with Rectified Flow Matching"
https://cantabile-kwok.github.io/VoiceFlow/
276 stars 20 forks source link

Aboute Dataset #12

Open mumuyeye opened 4 months ago

mumuyeye commented 4 months ago

Hello! I'm working on building my own dataset. Could you give me an example of Python code that generates phoneme sequences (text) and their corresponding durations (phn_duration).

cantabile-kwok commented 4 months ago

Hi, our data preprocessing was conducted long ago as a detached process, so I regret that I literally do not have a simple code example for this. But our processing was done via Kaldi, so you may find help from the most related tool MFA https://github.com/MontrealCorpusTools/Montreal-Forced-Aligner.