pipecat-plugins-faseeh

Pipecat plugin for Faseeh Arabic text-to-speech.

Build real-time Arabic conversational AI agents using Pipecat with Faseeh's high-quality Arabic TTS voices.

Installation

pip install pipecat-plugins-faseeh

Configuration

Set your Faseeh API key as an environment variable:

export FASEEH_API_KEY="your-api-key"

Or pass it directly:

tts = FaseehTTSService(api_key="your-api-key", aiohttp_session=session)

Get your API key at app.faseeh.ai.

Parameters

Usage

import aiohttp
from pipecat.pipeline.pipeline import Pipeline
from pipecat_plugins_faseeh import FaseehTTSService

async with aiohttp.ClientSession() as session:
    tts = FaseehTTSService(
        api_key="your-api-key",
        aiohttp_session=session,
        voice_id="ar-hijazi-female-2",
        model="faseeh-v1-preview",
        params=FaseehTTSService.InputParams(stability=0.5, speed=1.0),
    )

    # Add to your Pipecat pipeline
    pipeline = Pipeline([..., tts, ...])

Runtime settings update

Change voice, model, speed, or stability mid-conversation:

from pipecat.frames.frames import TTSUpdateSettingsFrame

await task.queue_frame(TTSUpdateSettingsFrame(settings={"voice_id": "ar-emirati-male-1"}))
await task.queue_frame(TTSUpdateSettingsFrame(settings={"speed": 1.1, "stability": 0.8}))

Running the example

# Install dependencies
pip install pipecat-plugins-faseeh "pipecat-ai[daily,openai,deepgram,silero]" python-dotenv loguru

# Copy and fill in your API keys
cp .env.example .env

# Run
python examples/faseeh_tts_example.py

The example auto-creates a Daily room and logs the URL — open it in your browser to talk to the bot.

Available voices

See the full voice library at docs.faseeh.ai.

Pipecat compatibility

Tested with pipecat-ai >= 0.0.100.

License

Apache 2.0

Maintained by

Actualize