OffLineVoiceProcessor

Piper TTS Setup – Local, Offline, Zero-Cloud Voice for Elite Intel

Elite Intel uses Piper / Piper TTS project for ultra-fast, fully-offline text-to-speech. No API keys, no Google – just pure local voice running on your own machine.

🐧 Linux Manual Install

## replace with whatever directory you want
sudo mkdir -p /path/to/piper-tts
cd /path/to/piper-tts

## install piper-tts
python3 -m venv .
source bin/activate
pip install --upgrade pip
pip install piper-tts[http]

## download voice models
## Amy
wget -O en_US-amy-medium.onnx https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/amy/medium/en_US-amy-medium.onnx
wget -O en_US-amy-medium.onnx.json https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/amy/medium/en_US-amy-medium.onnx.json

## Joe
wget -O en_US-joe-medium.onnx https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/joe/medium/en_US-joe-medium.onnx
wget -O en_US-joe-medium.onnx.json https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/joe/medium/en_US-joe-medium.onnx.json

Run Piper local Web server (still inside the virtual env)

python3 -m piper.http_server -m en_US-amy-medium.onnx

The server listens on port 5000.

To test, run

curl -X POST -H 'Content-Type: application/json' -d '{ "text": "This is a test." }' -o test.wav localhost:5000
aplay test.wav 

🪟 Windows Manual Install

1. Download the latest Windows build:
   https://github.com/rhasspy/piper/releases/latest
   → Pick one of these:

   • piper_windows_amd64.zip → CPU only (works everywhere)
   • piper_windows_amd64_cuda.zip → NVIDIA GPU (fastest)
   • piper_windows_amd64_directml.zip → AMD / Intel GPU

2. Extract the zip anywhere (e.g. C:\Piper\)

3. Put your voice file (.onnx + .onnx.json) in the same folder
   Recommended starter voice: en_US-amy-medium.onnx (included in most releases)

4. Start the HTTP server (this is what Elite Intel talks to):

   piper.exe --model en_US-amy-medium.onnx --port 5000

---

NOTE: In the app, just do not provide TTS key. And Piper will be used by default. Personalities and voice switching features are not supported. Defaults to "Professional / Federation"

👉You can download voices here👈