Nicheloom visits: 60,536 all-time · 7,913 in the last 7 days

The opportunity tracker for new startups.

Today: 895 launches · Niche of the day: browser-based video games · See what launched today →

tiny-voca

A 4.5M-parameter two-voice English text-to-speech model, trained from scratch

Get picks like this daily. Today's launch numbers, the top launches, AI and tech news, and a daily opportunity pick, straight to your inbox.

Want to reach these readers? Sponsor an issue.

GITHUB · launched Nov 14, 2026 · 354 upvotes (top 6%) · 0 views

Niche: speech and voice ai models · 145 launches, 9 in the last 7 days ▲ +13% · Explore this niche →

Category and tags

  • Horizontal
  • Model & infra
  • Developer
  • AI-native
  • Hobby / open-source project
  • onnx
  • text-to-speech
  • tts

Visit website ↗ Source listing ↗

Could you build this?

Partial A demo web interface for a small TTS model is easy to vibe-code, but training a 4.5M-parameter model from scratch needs ML expertise and a usable speech dataset that the builder must find or assemble.

What it would actually take: The stack is a small transformer or non-autoregressive TTS architecture trained in PyTorch, with a vocoder and a lightweight browser or server inference front end. The hard parts are curating clean multi-speaker speech data with transcripts and debugging training until output is intelligible. This takes ML engineering experience and GPU time, though the model is small enough to train on a single consumer or rented GPU.

Competitors

Other products that read as similar to this one : 1278 launches clear the similarity bar, closest 8 shown.

Attention rank: #101 of 1279 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).

Launched 380 days after the earliest competitor.

The niche, labels, vibe code rating and competitor comparison on this page are analysis by Nicheloom. If you use them, please credit nicheloom.com. How you may use our data.