Below agent code https://github.com/livekit/agents/blob/main/examples/other/text_only.py not working in livekit playground

Thanks for reaching out, Namit. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.

how to do last step

Also, confirm the agent is subscribed to the lk.chat topic to receive text input and is publishing to the lk.transcription topic for text output.

Hey there! To get text input and output working in the LiveKit Playground with the

text_only.py

agent, here’s what you need to do: 1. Start the agent in development mode by running:

python text_only.py dev

. 2. Open up the LiveKit Playground in your browser. If you’re using the hosted version, just head over to your configured playground URL. 3. Jump into the chat interface in the playground to send your text and see the agent’s responses. Just a quick reminder: make sure your environment variables (

LIVEKIT_API_KEY

,

LIVEKIT_API_SECRET

,

LIVEKIT_URL

) are all set up correctly so the agent can connect smoothly to the LiveKit server. Sources: Agents playground | LiveKit Docs | Text and transcriptions | LiveKit Docs | livekit-agents/livekit/agents/voice/chat_cli.py | .env.example

how to do last step

Also, confirm the agent is subscribed to the lk.chat topic to receive text input and is publishing to the lk.transcription topic for text output.

Let’s make sure your agent is all set up to subscribe to the

lk.chat

topic and publish to the

lk.transcription

topic. Here’s a quick checklist to help you out: 1. First, double-check that the agent is initialized with

RoomInputOptions(text_enabled=True)

. This turns on text input and subscribes to the

lk.chat

topic. 2. Next, confirm that

RoomOutputOptions(transcription_enabled=True)

is enabled so the agent can publish to the

lk.transcription

topic. 3. Take a look at the agent’s code to see if methods like

registerTextStreamHandler

for

lk.chat

and

publishTranscription

for

lk.transcription

are in place. 4. It’s also a good idea to turn on logging in the agent. This way, you can keep an eye on when messages come in on

lk.chat

and when transcriptions go out to

lk.transcription

. 5. On the frontend side, set up a text stream handler to check that messages are being received on both topics. Here’s a handy example:

room.registerTextStreamHandler('lk.transcription', async (reader, participantInfo) => {
    const message = await reader.readAll();
    console.log(`New message from ${participantInfo.identity}: ${message}`);
});

Following these steps should give you a clear picture of your agent’s subscription and publishing status. If you run into any bumps, I’m here to help! Sources: Text and transcriptions | LiveKit Docs | packages/core/src/observables/dataChannel.ts | agents/src/constants.ts

give above code in python

Here's a friendly example of how you can set up a text stream handler in Python for LiveKit to keep an eye on the

lk.chat

and

lk.transcription

topics:

from livekit import rtc

class TextStreamHandlerImpl:
    async def on_text_received(self, reader: rtc.TextStreamReader, participant_identity: str):
        """Handle incoming text streams for both lk.chat and lk.transcription topics"""
        async for chunk in reader:
            print(f"Received text chunk from {participant_identity}: {chunk}")
            # Feel free to add more processing here

async def main(room_name: str):
    # ... your existing connection setup ...

    # Let's create an instance of our handler
    text_handler = TextStreamHandlerImpl()

    # Now, we'll register handlers for both topics
    room.register_text_stream_handler(
        'lk.chat', 
        lambda reader, participant_identity: text_handler.on_text_received(reader, participant_identity)
    )
    room.register_text_stream_handler(
        'lk.transcription', 
        lambda reader, participant_identity: text_handler.on_text_received(reader, participant_identity)
    )

    # ... and then continue with the rest of your code ...

This snippet listens for text streams on both topics and prints out the messages it receives. You can easily build on this by adding whatever extra logic you need. Hope this helps! Sources: Text and transcriptions | LiveKit Docs | Text and transcriptions | LiveKit Docs | examples/other/text_only.py

RuntimeWarning: coroutine 'TextStreamHandlerImpl.on_text_received' was never awaited text_stream_handler(text_reader, participant_identity) RuntimeWarning: Enable tracemalloc to get the object allocation traceback

Hey there! The error popped up because the coroutine

on_text_received

wasn’t properly awaited. No worries though—just make sure all your asynchronous operations are awaited. Here’s a fixed version of the code for you:

import asyncio
from livekit import rtc

class TextStreamHandlerImpl:
    async def on_text_received(self, reader: rtc.TextStreamReader, participant_identity: str):
        try:
            async for chunk in reader:
                print(f"Received text chunk from {participant_identity}: {chunk}")
        except Exception as e:
            print(f"Error reading text stream: {e}")

async def main():
    # Initialize and connect to the room
    room = rtc.Room()
    # Assume you have your LiveKit URL and token
    await room.connect("Your LiveKit URL", "Your Token")

    # Create an instance of the handler
    text_handler = TextStreamHandlerImpl()

    # Register the text stream handler
    room.register_text_stream_handler(
        'lk.chat', 
        lambda reader, participant_identity: asyncio.create_task(text_handler.on_text_received(reader, participant_identity))
    )

    # Keep the connection alive
    while True:
        await asyncio.sleep(1)

if __name__ == "__main__":
    asyncio.run(main())

Here’s what we changed: • Used

asyncio.create_task

so the coroutine gets awaited properly. • Wrapped everything inside

asyncio.run

to handle the event loop smoothly. With these tweaks, you should be all set—no more runtime warnings, and your text streams will be handled just right. If you hit any other bumps, feel free to ask!