what if i write two generate repy method together example `` LiveKit Community #ask-ai

what if i write two generate_repy method together ...

colossal-xylophone-68609

04/15/2026, 6:03 PM

what if i write two generate_repy method together example

Copy code

await session.generate_reply("greet the user")
generate_reply("tell a joke to the user")

then will the agent execute the second generate reply instruction after the agent has said its full sentence for the first generate_reply call. or will it execute without waiting. or will it only wait till the generate_reply generates the message using llm and not wait for agent to fully complete saying it. what will happen here

✅ 1

tall-airline-16756

04/15/2026, 6:03 PM

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

tall-airline-16756

04/15/2026, 6:04 PM

Thanks — I’m answering your question about what happens when you call

generate_reply

twice. • If you

await

the first call (your example), the first reply will finish fully before the next call runs: a. LLM generates the first message b. the agent speaks the full first reply c. then the second

generate_reply()

executes • If you call both without

await

, they run concurrently and can overlap (generation and/or playback may interleave). • For finer control use the returned `SpeechHandle`: ◦

handle = session.generate_reply(...)

◦

await handle.wait_for_playout()

ensures playback finished before continuing. • Recommended for natural sequential speech: ◦

await session.generate_reply("greet the user")

◦

await session.generate_reply("tell a joke to the user")

If you want, tell me which behavior you need (strict sequencing vs. concurrent generation) and I’ll show the exact code pattern. Sources: Agent speech and audio | LiveKit Documentation, with interactive elements

2 Views

Open in Slack

Previous Next