livekit-plugins-induslabs 0.0.2

Agent Framework plugin for services using IndusLabs API.

IndusLabs plugin for LiveKit Agents

This repository contains a plugin that enables the LiveKit Agents framework to use IndusLabs for both text‑to‑speech (TTS) and speech‑to‑text (STT) services. By packaging the integration as a standard Python module you can install it with pip and immediately register it with your LiveKit agent.

Features

• Text‑to‑Speech (TTS) – synthesize natural‑sounding speech from text using IndusLabs models. Streaming support is provided so that audio begins playing as soon as the first bytes arrive.
• Speech‑to‑Text (STT) – transcribe speech to text in real time. The plugin uses a voice activity detector to segment utterances and sends them to the IndusLabs transcription API.
• Drop‑in installation – simply run pip install livekit‑plugins‑induslabs and import the plugin in your agent. The plugin automatically registers itself with the LiveKit plugin registry.

Installation

Ensure you have Python 3.8 or newer installed. Then run:

pip install livekit-plugins-induslabs

The package depends on the livekit-agents library and uses aiohttp, numpy and webrtcvad under the hood. These dependencies will be installed automatically.

You will also need an API key from IndusLabs. Set the INDUSLABS_API_KEY environment variable before using the plugin or pass the key explicitly when constructing the TTS class.

Usage

Import the plugin classes and pass them into your AgentSession:

from livekit.plugins.induslabs import TTS, STT

# Create instances (reads API key from INDUSLABS_API_KEY if not given)
tts = TTS(voice="Indus-hi-Urvashi")
stt = STT(sample_rate=16000)

session = AgentSession(
    stt=stt,
    tts=tts,
    # ... other components like llm, etc.
)

# run session...

Text‑to‑Speech streaming

response_stream = tts.stream()
response_stream.push("नमस्ते!")
async for event in response_stream:
    # Handle audio bytes from the emitter

Speech‑to‑Text streaming

speech_stream = stt.stream()
async for event in speech_stream:
    if event.type == SpeechEventType.FINAL_TRANSCRIPT:
        print("Final transcript:", event.alternatives[0].text)

Contributing

This project is provided as a reference integration for IndusLabs. If you encounter issues or have feature suggestions please open an issue or submit a pull request on GitHub.