pipecat/examples/video-avatar/video-avatar-tavus-video-service.py

#
# Copyright (c) 2024-2026, Daily
#
# SPDX-License-Identifier: BSD 2-Clause License
#


import datetime
import io
import os
import wave

import aiofiles
import aiohttp
from dotenv import load_dotenv
from loguru import logger

from pipecat.audio.vad.silero import SileroVADAnalyzer
from pipecat.frames.frames import LLMRunFrame
from pipecat.pipeline.pipeline import Pipeline
from pipecat.pipeline.runner import PipelineRunner
from pipecat.pipeline.task import PipelineParams, PipelineTask
from pipecat.processors.aggregators.llm_context import LLMContext
from pipecat.processors.aggregators.llm_response_universal import (
    LLMContextAggregatorPair,
    LLMUserAggregatorParams,
)
from pipecat.processors.audio.audio_buffer_processor import AudioBufferProcessor
from pipecat.runner.types import RunnerArguments
from pipecat.runner.utils import create_transport
from pipecat.services.cartesia.tts import CartesiaTTSService
from pipecat.services.deepgram.stt import DeepgramSTTService
from pipecat.services.google.llm import GoogleLLMService
from pipecat.services.tavus.video import TavusVideoService
from pipecat.transports.base_transport import BaseTransport, TransportParams
from pipecat.transports.daily.transport import DailyParams

load_dotenv(override=True)


async def save_audio_file(audio: bytes, filename: str, sample_rate: int, num_channels: int):
    """Save audio data to a WAV file."""
    if len(audio) > 0:
        with io.BytesIO() as buffer:
            with wave.open(buffer, "wb") as wf:
                wf.setsampwidth(2)
                wf.setnchannels(num_channels)
                wf.setframerate(sample_rate)
                wf.writeframes(audio)
            async with aiofiles.open(filename, "wb") as file:
                await file.write(buffer.getvalue())
        logger.info(f"Audio saved to {filename}")


# We use lambdas to defer transport parameter creation until the transport
# type is selected at runtime.
transport_params = {
    "daily": lambda: DailyParams(
        audio_in_enabled=True,
        audio_out_enabled=True,
        video_out_enabled=True,
        video_out_is_live=True,
        video_out_width=1280,
        video_out_height=720,
        audio_out_auto_silence=False,
    ),
    "webrtc": lambda: TransportParams(
        audio_in_enabled=True,
        audio_out_enabled=True,
        video_out_enabled=True,
        video_out_is_live=True,
        video_out_width=1280,
        video_out_height=720,
        audio_out_auto_silence=False,
    ),
}


async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
    logger.info(f"Starting bot")
    async with aiohttp.ClientSession() as session:
        stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"], audio_passthrough=True)

        tts = CartesiaTTSService(
            api_key=os.environ["CARTESIA_API_KEY"],
            settings=CartesiaTTSService.Settings(
                voice="a167e0f3-df7e-4d52-a9c3-f949145efdab",
            ),
        )

        llm = GoogleLLMService(
            api_key=os.environ["GOOGLE_API_KEY"],
            settings=GoogleLLMService.Settings(
                system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
            ),
        )

        tavus = TavusVideoService(
            api_key=os.environ["TAVUS_API_KEY"],
            replica_id=os.environ["TAVUS_REPLICA_ID"],
            session=session,
        )

        context = LLMContext()
        user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
            context,
            user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
        )

        audiobuffer = AudioBufferProcessor()

        pipeline = Pipeline(
            [
                transport.input(),  # Transport user input
                stt,  # STT
                user_aggregator,  # User responses
                llm,  # LLM
                tts,  # TTS
                tavus,  # Tavus output layer
                transport.output(),  # Transport bot output
                audiobuffer,  # Audio recording
                assistant_aggregator,  # Assistant spoken responses
            ]
        )

        task = PipelineTask(
            pipeline,
            params=PipelineParams(
                audio_in_sample_rate=16000,
                audio_out_sample_rate=24000,
                enable_metrics=True,
                enable_usage_metrics=True,
            ),
            idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
        )

        @transport.event_handler("on_client_connected")
        async def on_client_connected(transport, client):
            logger.info(f"Client connected")
            await audiobuffer.start_recording()
            # Kick off the conversation.
            context.add_message(
                {
                    "role": "user",
                    "content": "Start by greeting the user and ask how you can help.",
                }
            )
            await task.queue_frames([LLMRunFrame()])

        @transport.event_handler("on_client_disconnected")
        async def on_client_disconnected(transport, client):
            logger.info(f"Client disconnected")
            await task.cancel()

        @audiobuffer.event_handler("on_audio_data")
        async def on_audio_data(buffer, audio, sample_rate, num_channels):
            timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
            filename = f"recordings/merged_{timestamp}.wav"
            os.makedirs("recordings", exist_ok=True)
            await save_audio_file(audio, filename, sample_rate, num_channels)

        @audiobuffer.event_handler("on_track_audio_data")
        async def on_track_audio_data(buffer, user_audio, bot_audio, sample_rate, num_channels):
            timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
            os.makedirs("recordings", exist_ok=True)
            await save_audio_file(user_audio, f"recordings/user_{timestamp}.wav", sample_rate, 1)
            await save_audio_file(bot_audio, f"recordings/bot_{timestamp}.wav", sample_rate, 1)

        runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)

        await runner.run(task)


async def bot(runner_args: RunnerArguments):
    """Main bot entry point compatible with Pipecat Cloud."""
    transport = await create_transport(runner_args, transport_params)
    await run_bot(transport, runner_args)


if __name__ == "__main__":
    from pipecat.runner.run import main

    main()