🛡️ Privacy-focused user? Please check our Documentation before downloading.

---

🏆 Featured In

VoxSherpa TTS is listed in the official README of k2-fsa/sherpa-onnx — the core inference library powering this app.

---

Why VoxSherpa?

Most TTS apps make you choose between quality and privacy. Cloud-based tools like ElevenLabs sound incredible — but they require internet, send your text to remote servers, and charge per character.

VoxSherpa breaks that tradeoff.

It runs two professional-grade neural engines entirely on your device:

Engine	Quality	Speed	Best For
🧠 Kokoro-82M	Studio-grade · rivals ElevenLabs	Slower on budget hardware	Audiobooks, voiceovers, professional content
⚡ Piper / VITS	Natural · clear · multi-speaker	Fast on any device	Daily use, dialogue synthesis, quick synthesis

---

Screenshots

---

Features

🎙️ Dual Neural Engine

• Kokoro-82M — 82 million parameter neural model. Multilingual support including Hindi, English, British English, French, Spanish, Chinese, Japanese and 50+ languages. Same architecture used by top-tier commercial TTS services.
• Piper / VITS — Fast, lightweight, natural. Generates speech in seconds on any Android device.

🗣️ Multi-Speaker Dialogue Synthesis (New in v4.0)

• Piper now supports multiple speakers within a single script — generate full conversations in one pass instead of stitching clips together
• New [speaker] tag system to assign each line to a distinct voice directly inside your script:

  [speaker:1] Hello, how are you?
  [speaker:2] I'm good, thanks! How about you?
  

• Voice Style & Tone controls — fine-tune how each speaker sounds to match the mood of the script
• Adjustable sentence gap / silence timing — control the pause duration between lines for natural, realistic pacing
• Ideal for audiobooks with multiple characters, podcasts, skits, and narrated dialogue

🔒 100% Offline & Private

• All processing happens on your device
• No internet required after model download
• No account, no telemetry, no data collection
• Your text never leaves your phone

📄 Document to Audio

• PDF to Audio — listen to any document hands-free
• TXT to Audio — convert plain text files instantly
• Share any text directly to VoxSherpa from any app

📦 Model Management

• Download models directly from the app
• Filter voice models by language or type
• Sample voice preview before selecting a model
• Import your own .onnx models from local storage
• Multiple models installed simultaneously
• Smart storage tracking
• Optional MMS model support — enable via a toggle in Settings if needed

🔊 System-Wide TTS

• Set VoxSherpa as your default Android TTS engine
• All downloaded models exposed to System TTS — use any voice in Chrome, WhatsApp, TalkBack, and more
• Pitch & speed control in System TTS mode
• Sample voice preview for all models

🎧 Audio Controls

• Real-time waveform visualization
• Adjustable speed and pitch
• Interactive audio seeking with mini player controls
• MediaStyle notification with full playback controls
• Export as WAV with correct sample rate per model

📚 Speech Library

• Save all generated audio locally
• Favorites system for quick access
• View generation history with timestamps
• Voice model attribution per recording
• Regenerate audio on voice change

⚙️ Smart Settings

• Smart Punctuation — natural pauses after sentence breaks
• Emotion Tags — [whisper], [angry], [happy] support
• Per-model voice selection (Kokoro supports 50+ speakers)
• Theme-aware UI

---

Technical Architecture

User Text
    │
    ├─── Kokoro Engine (KokoroEngine.java)
    │         └── Sherpa-ONNX JNI → ONNX Runtime → CPU/NNAPI
    │                   └── kokoro-multi-lang-v1_0
    │
    └─── Piper / VITS Engine (VoiceEngine.java)
              └── Sherpa-ONNX JNI → ONNX Runtime → CPU
                      └── VITS model (language-specific)

Built with:

• Sherpa-ONNX — on-device neural inference
• Kokoro-82M — multilingual neural TTS model
• Piper — fast local TTS, now with multi-speaker dialogue support
• Android AudioTrack API — low-latency PCM playback

---

Performance

Generation speed depends entirely on your device's processor:

Device Tier	Kokoro	Piper
🟢 Flagship (Snapdragon 8 Gen 3)	~20–40 sec/min audio	~5 sec/min audio
🟡 Mid-range (8-core)	~60–90 sec/min audio	~10 sec/min audio
🔴 Budget (6-core)	~2–3 min/min audio	~20 sec/min audio

Kokoro prioritizes quality over speed by design. It uses the same 82M parameter architecture that powers premium commercial TTS — running it entirely offline on a mobile CPU is genuinely pushing the hardware limits.

---

Installation

Requirements: Android 11+ · ARM64 · ~500 MB free storage recommended (for models)

---

Model Import (Technical Users)

VoxSherpa supports importing custom .onnx models without any server:

1. Place your .onnx model + tokens.txt on device storage
2. Open Models tab → tap + → Import Local Model
3. Select your files

Compatible with any Sherpa-ONNX compatible TTS model.

---

Contributing

VoxSherpa is open source. Contributions welcome:

• 🐛 Bug reports via Issues
• 💡 Feature requests via Discussions
• 🔧 Pull requests for fixes and improvements

---

License

Copyright (C) 2025 CodeBySonu95

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

https://www.gnu.org/licenses/gpl-3.0.html

---

Acknowledgements

• k2-fsa/sherpa-onnx — the inference engine that makes this possible
• hexgrad/Kokoro-82M — the neural model behind studio-quality synthesis
• rhasspy/piper — fast local TTS engine, now with multi-speaker dialogue support

---