Pipecat Lokutor TTS

Lokutor text-to-speech integration for Pipecat.

Get an API key and test voices at app.lokutor.com. Documentation: docs.lokutor.com

Lokutor TTS runs entirely on CPU-only infrastructure, keeping costs low while delivering high-quality speech synthesis. No GPU markup — just affordable, scalable TTS.

Installation

pip install pipecat-lokutor

Or with uv:

uv add pipecat-lokutor

Usage

from pipecat_lokutor import LokutorTTSService

tts = LokutorTTSService(
    api_key="your_api_key",
    voice_id="F1",
    params=LokutorTTSService.InputParams(
        language="en",
        speed=1.0,
        steps=5,
        visemes=False,
    ),
)

Pipecat Pipeline

from pipecat.pipeline.pipeline import Pipeline

pipeline = Pipeline([
    transport.input(),
    stt,
    llm,
    tts,
    transport.output(),
])

Voices

Browse and test available voices at app.lokutor.com.

Supported Languages

32 languages: en, es, fr, de, it, pt, ja, ko, zh, ar, bg, hr, cs, da, nl, et, fi, el, hi, hu, id, lv, lt, pl, ro, ru, sk, sl, sv, tr, uk, vi.

Pass any Pipecat Language (including regional variants like Language.EN_US) — it is normalized to the matching Lokutor base code automatically.

Example

See examples/groq-stt-groq-llm-lokutor-tts.py for a full example using Groq STT + Groq LLM + Lokutor TTS.

# Install dependencies
uv sync

# Set environment variables
export GROQ_API_KEY="your_groq_api_key"
export LOKUTOR_API_KEY="your_lokutor_api_key"

# Run the example
python examples/groq-stt-groq-llm-lokutor-tts.py -t webrtc

Then open http://localhost:7860/client/.

Compatibility

Tested with Pipecat v0.0.86+.

License

BSD 2-Clause License. See LICENSE for details.