Google has announced that, starting in 2026/2027, all apps on certified Android devices
will require the developer to submit personal identity details directly to Google.
Since the developers of this app do not agree to this requirement, this app will no longer 
work on certified Android devices after that time.

Donate

Send a coffee to 
woheller69@t-online.de 


  
Or via this link (with fees)

SherpaTTS

SherpaTTS is an Android Text-to-Speech engine based on Next-gen Kaldi. It uses voices from Piper Voices or Coqui.

Installing Voice Models

From Hugging Face

The app provides a built-in model downloader that lets you install Piper and Coqui voice models directly from the "Manage Languages" screen. Simply select a language and the app will download and install the model automatically. Please note that this is the only instance where internet permission is required. Once the model is downloaded, text-to-speech works entirely offline, ensuring your privacy and convenience. Voices can be tested here.

From SD Card (Sideloading)

You can install custom Piper models directly from your device storage:

1. Open the app and go to "Manage Languages"
2. Tap "Install from SD Card"
3. Enter a 3-letter language code (e.g., eng, deu, fra)
4. Enter a model name (this will be displayed in the language list)
5. Select your model file (.onnx format)
6. Select your tokens file (.txt format)
7. Tap "OK"

Requirements:

• The model file must be in .onnx format (make sure the model has been converted a described here: Sherpa ONNX conversion)
• The tokens file must contain the vocabulary mapping (typically tokens.txt)

Via ADB (Alternative Method)

You can also install models via ADB:

1. Create a directory: modelDir in sdcard/Android/data/org.woheller69.ttsengine/files
2. Place 3 files there:
   • model.onnx - The converted model file
   • tokens.txt - The vocabulary file
   • lang - A text file with:
     • Line 1: 3-letter language code (e.g., eng)
     • Line 2: Model name

At next start, the app will migrate it to the new directory structure and add it to installed languages.

Make sure the model has been converted a described here: Sherpa ONNX conversion

Contribute

For translations use https://toolate.othing.xyz/projects/sherpatts/

License

This work is licensed under GPLv3 license, © woheller69

• This app is based on the Sherpa ONNX Project, published under Apache-2.0 license
• It uses data from eSpeak NG, published under GPLv3 license
• It uses jsoup, published under MIT license
• At first start it downloads and installs a Piper or Coqui voice model from Hugging Face.

OTHER APPS

RadarWeather	Gas Prices	Smart Eggtimer
Bubble	hEARtest	GPS Cockpit
Audio Analyzer	LavSeeker	TimeLapseCam
Arity	Cirrus	solXpect
gptAssist	dumpSeeker	huggingAssist
FREE Browser	whoBIRD	PeakOrama
Whisper	Seamless	SherpaTTS