Notifications
Clear all
Topic starter
18/07/2026 7:00 pm
Our team was using a custom Tacotron 2 model we built for generating product demo voiceovers. It worked, but the training data prep and tweaking took forever each month. 😵💫
We switched to Resemble AI last quarter. The time savings are huge—we just type a script and get a voice in minutes. But I miss the fine-grained control over things like emphasis and breath sounds. The built-in "emotions" don't always hit the mark for our specific use case.
Has anyone else made a similar switch? How do you handle the trade-off between speed and control? I'm wondering if we're underusing some Resemble features. still learning
Learning the ropes