Termux-TTS (Python)

High-Performance Multi-Backend Speech Synthesis Framework for Android Termux & Edge Linux
Supports Zero-Dependency Parametric DSP Formant Vocoder, Deep Learning ONNX Neural Runtime, and Android Native System Voice Bridge.

Installation

pip install termux-tts

Quickstart

import termux_tts as tts

# 1. Zero-Dependency DSP Formant Mode (Default)
with tts.load(engine="dsp", language="ko") as engine:
    res = engine.synthesize("안녕하세요, 텀묵스 음성 합성입니다.", output="dsp.wav")
    print(f"Synthesized in {res.elapsed_ms:.1f}ms (RTF: {res.rtf:.4f}x)")

# 2. Deep Learning ONNX Mode
with tts.load(engine="onnx", model="vits_ko.onnx", language="ko") as engine:
    res = engine.synthesize("신경망 고품질 음성 합성입니다.", output="onnx.wav")

# 3. Android System Native Speak Mode
with tts.load(engine="native", language="ko") as engine:
    engine.speak("스피커로 즉시 발화합니다.")

Features

Zero-Dependency DSP Engine: Instant CPU execution with 0MB download overhead (Rosenberg Glottal Pulse + 5-Band Biquad Filters).
Authentic ONNX Neural Runtime: Deep learning acoustic model execution for VITS, Piper, and FastSpeech ONNX models.
Android Native Bridge: Direct speaker speech playback via Termux IPC.
Cross-Platform CLI & Node.js Binding: First-class support for Python, npm, and shell scripts.

Documentation

License

Release files for termux-tts 1.1.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distribution (wheel)

Table of built distributions (wheels) for termux-tts 1.1.2
File	Interpreter	ABI	Platform	Reset
termux_tts-1.1.2-py3-none-any.whl 23.4 kB Sep 2, 2026	Python 3	none	any	Details

Release files / termux_tts-1.1.2-py3-none-any.whl

Download URL	termux_tts-1.1.2-py3-none-any.whl
Size	23.4 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`e8cd2d110173285470ffc02b8e108847db587f81768e55ab8e033154025f0e29`
BLAKE2b-256 checksum How to use checksums	`9368dec1114371e6910675755b0501c1a85d7b0417084f081e2827c5e1634eb0`
Upload date	Sep 2, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.12.0`