Skip to content

VoxSherpa TTS v1.0-beta

Pre-release
Pre-release

Choose a tag to compare

View all tags
@CodeBySonu95 CodeBySonu95 released this 20 Mar 11:44
· 117 commits to main since this release
v1.0-beta
643a046
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

🎙️ VoxSherpa TTS — v1.0-beta

First public pre-release. Core features are stable and ready for testing.

What's New

This is the first public release of VoxSherpa TTS — a fully offline neural text-to-speech app for Android.

Engines

  • Kokoro-82M — studio-quality multilingual TTS (Hindi, English, 50+ languages)
  • Piper / VITS — fast, lightweight neural TTS for everyday use

Features

  • Download and manage voice models directly from the app
  • Import local .onnx models from device storage
  • Real-time waveform visualization
  • Save generated audio as WAV files
  • Speech history library with favorites
  • Adjustable speed and pitch
  • Smart Punctuation & Emotion Tags support
  • 100% offline — no internet required after model download

Known Limitations

  • Kokoro generation is slow on budget/mid-range devices (2–3 min per minute of audio) — this is a hardware limitation, not a bug
  • Firebase-based online model list requires internet for first-time model discovery (local import works fully offline)

Requirements

  • Android 11 (API 30) or higher
  • ~500 MB free storage recommended (for models)

Installation

  1. Download VoxSherpa-TTS-v1.0-beta.apk below
  2. Enable Install from unknown sources in Android settings
  3. Install and open the app
  4. Go to Models tab → Download a voice model → Start generating

Feedback

Found a bug or have a suggestion? Open an Issue on GitHub.

⚠️ This is a pre-release. Expect rough edges. Production release coming soon on Play Store

Assets 3
Loading