TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: TTS Accents – 1000 voice combinations

4 pointsby yoeven5 months ago
Most TTS models today have a fixed set of voices that are tied to a fixed set of languages.<p>We built a trained TTS model with the concept of accents like we have in real life. You can have a Chinese person speaking English or a German person speaking French. Hints of their accents from their original language can be heard when they speak another language.<p>We took this idea and the understanding of how one-shot voice cloning works and trained a model on 100+ voices &amp; languages, giving you 1000+ unique voice combinations.<p>One interesting use case is customer support to have a voice sound similar to where the person is calling from, making them feel more at home!<p>It&#x27;s still pretty new, and there&#x27;s more to come, but we do have users using it in production, open to any feedback and ideas :)<p>Try it for free: <a href="https:&#x2F;&#x2F;jigsawstack.com&#x2F;text-to-speech" rel="nofollow">https:&#x2F;&#x2F;jigsawstack.com&#x2F;text-to-speech</a>

1 comment

itake5 months ago
What data did you use to train this?
评论 #42403217 未加载