4 ms·
What none of the text to speech generators seem to get right is — the aspects that make real human podcasts easier to listen to: hesitations, rephrasing, pauses
by d4rkp4ttern 4y ago
What none of the text to speech generators seem to get right is — the aspects that make real human podcasts easier to listen to: hesitations, rephrasing, pauses, variation in speed, intonation etc.
I have yet to see something like this. Something less “perfect” sounding than say the google maps voice.
- bacchusracine 4y ago>the aspects that make real human podcasts easier to listen to: hesitations, rephrasing, pauses, variation in speed, intonation etc. Have you heard ShowDoJo's Wurst Take yet? https://www.twitch.tv/showdojo https://www.twitch.tv/showdojo It's not perfect but it's one of the best I've heard so far.