SSML has some support for changing prosody, it would be fun to support at least one method. So with this in mind look at implementing TD-PSOLA just for simple time shifting (and also unlocks rudimentary pitch shifting)
https://speech.zone/td-psola-the-hard-way/
https://speech.zone/courses/speech-processing/module-6-speech-synthesis-waveform-generation-and-connected-speech/videos/pitch-period/
https://crates.io/crates/pitch-detection
SSML has some support for changing prosody, it would be fun to support at least one method. So with this in mind look at implementing TD-PSOLA just for simple time shifting (and also unlocks rudimentary pitch shifting)
https://speech.zone/td-psola-the-hard-way/
https://speech.zone/courses/speech-processing/module-6-speech-synthesis-waveform-generation-and-connected-speech/videos/pitch-period/
https://crates.io/crates/pitch-detection