# Kowairo Voice — On-Device Japanese TTS

> On-Device Japanese TTS (iPhone/iPad app by TAKAYOSHI YOSHIKAWA.)

- Source: https://appshunter.io/ios/app/kowairo-voice/id6782555687 (this page in markdown: same URL + `.md`)
- Developer: [TAKAYOSHI YOSHIKAWA](https://appshunter.io/developer/1882343499)
- Category: Utilities
- Price: Free
- Age rating: 4+
- Requires: iOS 18.0 · 35 MB
- Languages: English
- Released: 2026-07-06
- Data updated: 2026-08-10
- User reviews in markdown: https://appshunter.io/ios/app/kowairo-voice/id6782555687/reviews.md

## What is Kowairo Voice?

Kowairo Voice is a completely offline Japanese text-to-speech app powered by Irodori-TTS v3. It runs entirely on-device — no internet connection required after the initial model download.

FULLY OFFLINE & PRIVATE
All processing happens on your device. Your text and generated audio never leave your phone. No data collection, no accounts needed.

VOICE DESIGN
Describe the voice you want in natural language — "calm female voice, cheerful" or "energetic young male voice" — and the AI generates speech matching your description.

VOICE CLONING (DUAL MODE)
Upload a reference audio recording to clone the voice. Combine with text descriptions for even more control.

ORGANIZED LIBRARY
Browse your generated clips by date, search by text content, and mark favorites for quick access.

PLAYLISTS
Group your audio clips into playlists. Play them in order, shuffled, or on repeat.

POWERED BY IRODORI-TTS V3
Built on the latest Irodori-TTS v3 (VoiceDesign) model, fine-tuned for high-quality Japanese speech synthesis with expressive control.

Perfect for content creators, language learners, accessibility needs, and anyone who wants to generate natural Japanese speech on their own terms.


## Version history (last 4 releases)

### 1.3 — 2026-08-10

- You can now choose the voice model quality
- Long-form batch generation is smoother
- Refreshed the first-time guide

Everything still runs entirely on your device.

### 1.2 — 2026-07-24

An update focused on everyday usability.

- Audio you save to a playlist or favorites is now excluded from the library's automatic cleanup, so you can keep what matters without worrying about limits
- Audio added to a playlist is automatically starred, so you can see at a glance what's kept
- Improved long-form auto-splitting: blank lines are treated as paragraph breaks, and runs of emoji or symbols are no longer split in the middle
- Reorganized the Settings screen so options are easier to find

Still fully offline. Your text and audio never leave the device.

### 1.1 — 2026-07-08

Added long-form batch generation. Paste a long script and the app automatically splits it into well-sized blocks and synthesizes them together. With voice cloning, the tone stays consistent across chapters.

- New long-form batch generation mode
- Splits naturally at sentence boundaries ("。", line breaks, etc.)
- Optimized block length so long texts stay smooth
- Faster generation to cut waiting time on long texts
- Smoother tap-to-edit for blocks, with fine volume / L-R balance control

Still fully offline. Your text and audio never leave the device.

### 1.0 — 2026-07-06

No release notes.

## More apps by TAKAYOSHI YOSHIKAWA

- [Still Blinking](https://appshunter.io/ios/app/still-blinking/id6760018994)
- [PivotClock](https://appshunter.io/ios/app/pivotclock/id6760578309)
- [My Sound Alarm](https://appshunter.io/ios/app/my-sound-alarm/id6760616800)
- [RecExport: Voice to Cloud](https://appshunter.io/ios/app/recexport-voice-to-cloud/id6761234228)

All apps by TAKAYOSHI YOSHIKAWA: https://appshunter.io/developer/1882343499

---

*Data collected daily from the US App Store and indexed by [AppsHunter](https://appshunter.io/). User reviews are verbatim App Store reviews. Ratings, prices and chart positions refresh continuously; this snapshot is from 2026-08-10.*