RealtimeTTS

RealtimeTTS is a Python text-to-speech library for applications that need to turn strings, generators, and LLM token streams into audio with low latency. It can play speech locally, stream chunks to another process, write WAV files, and fall back across multiple engines.

The project supports a broad engine matrix: local system voices, cloud APIs, free service wrappers, local neural models, and voice-cloning stacks.

Support RealtimeTTS

If RealtimeTTS saved you time, one GitHub star is a simple way to help make it more stable.

Stars improve visibility, and visibility brings more users, more real-world testing, more bug reports, more fixes, and better releases for everyone.

Demo

Short_RealtimeTTS_Demo.mov

Recommended Engine: QwenEngine

For supported Windows and Linux systems with an NVIDIA GPU, QwenEngine is currently the recommended and preferred RealtimeTTS engine for high-quality, low-latency conversational speech. It offers multilingual Qwen3-TTS quality, x-vector and ICL voice cloning, native 24 kHz PCM streaming, fast cancellation, and the same engine either in-process or behind the production Qwen server.

In 10 warm Linux runs on our tuned RTX 4090 setup, the timeline was: about 35 ms engine TTFT, another 35 ms until RealtimeTTS emits its first PCM chunk, and about 10 ms of silence inside that chunk. Predicted audible onset was 80.9 ms and RTF was 0.108. These are orientation figures; measure the complete path on your target system.

python -m pip install --only-binary=realtimetts-qwen-native "realtimetts[qwen]"
python -m qwentts_cpp doctor

See the QwenEngine guide for setup and details. InflectEngine is the documented lightweight alternative for one fixed English voice on CUDA or ONNX CPU.

Install

For the fastest local smoke test, install the system engine:

pip install "realtimetts[system]"

The system and other traditional engine extras use PyAudio. On Linux, install PortAudio headers before those extras:

sudo apt-get update
sudo apt-get install python3-dev portaudio19-dev

On macOS:

brew install portaudio

The native qwen and Inflect extras use the established PyAudio/PortAudio playback path. Windows has prebuilt PyAudio wheels for Python 3.10–3.13. On Linux and macOS, install PortAudio first using the commands above. The Qwen native wheel itself does not require a local CUDA Toolkit. RealtimeTTS 0.7.4 declares validated native Qwen wheels only for x86-64 Windows and Linux; macOS and other platforms are not supported release targets.

Install realtimetts[qwen-server] to expose the same native engine through an OpenAI-compatible HTTP API. The server provides /v1/audio/speech, dynamic voice registration, persistent voice latents, and watchdog-ready request/stall metrics on /health; it is headless and does not install PyAudio/PortAudio. The server defaults to loopback (127.0.0.1). LAN exposure requires a deliberate --allow-lan bind plus a built-in API key, or a trusted reverse proxy that terminates TLS and enforces authentication. CORS defaults to explicit localhost origins and rejects wildcard *; CORS is not an access control boundary. See the Qwen guide for deployment, protocol, licensing, and asset boundaries.

Sentence splitting defaults to stream2sentence's nltk+rule-based consensus mode. The normal install, including realtimetts[qwen], installs stream2sentence[nltk] but not Stanza or PyTorch. Add Stanza only when wanted:

pip install "realtimetts[stanza]"
# or
pip install "realtimetts[qwen,stanza]"

For cloud engines, local neural engines, CUDA, mpv, and current packaging caveats, see docs/installation.md.

First Audio

from RealtimeTTS import TextToAudioStream, SystemEngine


if __name__ == "__main__":
    stream = TextToAudioStream(SystemEngine())
    stream.feed("Hello from RealtimeTTS.")
    stream.play()

Use the if __name__ == "__main__": guard in scripts, especially on Windows and when using engines that start worker processes.

Streaming Text

feed() accepts an iterator, so text can arrive while audio is already playing:

from RealtimeTTS import TextToAudioStream, SystemEngine


def text_chunks():
    yield "This starts speaking quickly. "
    yield "More text can arrive while audio is already playing."


if __name__ == "__main__":
    stream = TextToAudioStream(SystemEngine())
    stream.feed(text_chunks())
    stream.play()

Use the same pattern with an LLM client by yielding only non-empty text chunks. See docs/llm-streaming.md.

Output

Write audio to a WAV file without local speaker playback:

from RealtimeTTS import TextToAudioStream, SystemEngine


if __name__ == "__main__":
    stream = TextToAudioStream(SystemEngine())
    stream.feed("Save this speech to a file.")
    stream.play(output_wavfile="speech.wav", muted=True)

For output devices, mpv playback, muted mode, callbacks, and chunk formats, see docs/output-and-files.md.

Features

• Low-latency playback from strings, generators, and streamed model output.
• Multiple engines with local, cloud, free-service, and neural model options.
• Fallback engines for more resilient synthesis.
• Sync and async playback with pause, resume, stop, and state inspection.
• Text, audio, sentence, character, word-timing, and audio-chunk callbacks.
• WAV output, muted synthesis, selected output devices, and volume control.
• Voice switching and voice-cloning workflows where supported by the engine.

Engine Overview

Engine	Type	Install/status note	Best first use
QwenEngine (recommended)	Local native neural / HTTP server	realtimetts[qwen] or realtimetts[qwen-server] with a matching native wheel	High-quality multilingual realtime speech, voice cloning, and fast cancellation.
InflectEngine	Local lightweight	realtimetts[inflect]	Fast fixed English voice through PyTorch CUDA or ONNX CPU.
SystemEngine	Local	realtimetts[system]	First local audio smoke test.
GTTSEngine	Free service	realtimetts[gtts]	Simple network-backed speech.
EdgeEngine	Free service	realtimetts[edge], needs mpv	Free streamed voices.
OpenAIEngine	Cloud API	realtimetts[openai]	OpenAI TTS voices.
AzureEngine	Cloud API	realtimetts[azure]	Azure voices and word timings.
ElevenlabsEngine	Cloud API	realtimetts[elevenlabs], needs mpv	High-quality API voices.
CambEngine	Cloud API	realtimetts[camb]	CAMB MARS API voices.
MiniMaxEngine	Cloud API	realtimetts[minimax]	MiniMax cloud voices.
CartesiaEngine	Cloud API	realtimetts[cartesia]	Cartesia API voices.
TypecastEngine	Cloud API	realtimetts[typecast]	Typecast API voices.
ModelsLabEngine	Cloud API	realtimetts[modelslab]	ModelsLab API voices.
CoquiEngine	Local neural	realtimetts[coqui]	Local XTTS voice cloning.
PiperEngine	Local executable	realtimetts[piper], external Piper setup	Fast local executable TTS.
StyleTTSEngine	Local neural	realtimetts[styletts], local checkout/assets	StyleTTS experiments.
ParlerEngine	Local neural	realtimetts[parler]	GPU local model experiments.
KokoroEngine	Local neural	realtimetts[kokoro]	Local voices and timing support.
OrpheusEngine	Local/API-style	realtimetts[orpheus]	Orpheus model workflows.
OmniVoiceEngine	Local neural	realtimetts[omnivoice]	Multilingual voice cloning.
PocketTTSEngine / PocketTTSGpuEngine	Local lightweight	realtimetts[pockettts], realtimetts[pockettts-gpu] plus GPU fork	CPU-oriented voice cloning, optional CUDA fork path.
NeuTTSEngine	Local neural	realtimetts[neutts], optional neutts-gguf	Reference-audio voice cloning.
ZipVoiceEngine	Local neural	realtimetts[zipvoice], external checkout	ZipVoice cloning/server demos.
LuxTTSEngine	Local neural	realtimetts[luxtts]	LuxTTS voice cloning.
ChatterboxEngine	Local neural	realtimetts[chatterbox]	Chatterbox prompt-audio voices.
SoproTTSEngine	Local neural	realtimetts[sopro]	Sopro reference-audio voices.
SopranoEngine	Local neural	realtimetts[soprano]	Soprano local synthesis.
MossTTSEngine	Local neural	realtimetts[moss], runtime assets	MOSS-TTS experiments.

See docs/engine-selection.md before choosing an engine for an application. The engine-specific docs are being split out from the old README and source audit.

Documentation

• Quick start: shortest working examples.
• Installation: extras, platform setup, external tools, API keys, and known packaging mismatches.
• Engine selection: engine matrix and selection guidance.
• Feed and playback: feed(), play(), play_async(), pause, resume, stop, text state, and inline tags.
• LLM streaming: provider-neutral streamed text patterns and latency tuning.
• Output and files: WAV files, audio chunks, muted mode, output devices, mpv, buffering, and volume.
• Engine setup pages now link one focused page for each concrete engine source.
• FAQ: legacy troubleshooting page while topic docs are being split out.

Legacy translated docs remain under docs/<locale>/ while English is refactored as the canonical source.

Server Example

The browser and WebSocket server example lives in example_fast_api/:

python -m pip install fastapi uvicorn websockets pyaudio
python example_fast_api/async_server.py

Open http://localhost:8000 or connect to ws://localhost:8000/ws.

Related Project

RealtimeSTT is the speech-to-text counterpart for realtime voice input.

Contributing

Focused docs, tests, and engine fixes are easiest to review. During the docs refactor, keep English docs canonical and note mismatches between source, packaging, examples, and tests rather than hiding them.

License

RealtimeTTS source code is MIT licensed. Engine providers, model weights, voice data, datasets, generated audio, and third-party services can have separate terms. Read LICENSING_ADDENDUM.md and the relevant provider or model licenses before commercial use.

For the native Qwen/Inflect paths, qwentts.cpp and realtimetts-qwen-native are MIT-licensed; Qwen 0.6B Base/tokenizer and Inflect Micro-v2/ONNX are Apache-2.0. Model weights, voice latents, and reference audio are not bundled, and users are responsible for the rights to every voice or recording they provide.

Audio samples derived from the EARS dataset by Meta are licensed under CC BY-NC 4.0. See the original dataset terms for details.

Author

Kolja Beigel