When i have 2 participants connected into a single...
# agents
k
When i have 2 participants connected into a single room. only 1st participant's speech is processing by speech engine(stt) in that room. is it possible to process both of the participants speech into text and add in to transcript? if so, Please let me know how to do that using nodejs
l
We had a similar question yesterday, I presume that is coincidence, we have an example that shows something similar in Python but not Node.js, though the principle should be the same. https://livekit-users.slack.com/archives/C07FY8WHGPM/p1763452717363119
p
Related I am working on something similar. Multiple participants and the agents transcribes their audio. This works with the above python code but when I decided to port to use the node version, only the first participant audio gets transcribed. I get an error when an participant joins the room
Copy code
err: {
      "type": "Error",
      "message": "no job context found, are you running this code inside a job entrypoint?",
      "stack":
          Error: no job context found, are you running this code inside a job entrypoint?
              at getJobContext (/Users/mac/Documents/side/transcriber-node/node_modules/.pnpm/@livekit+agents@1.0.19_@livekit+rtc-node@0.13.21_zod@3.25.76/node_modules/@livekit/agents/src/job.ts:31:11)
              at AgentSession.start (/Users/mac/Documents/side/transcriber-node/node_modules/.pnpm/@livekit+agents@1.0.19_@livekit+rtc-node@0.13.21_zod@3.25.76/node_modules/@livekit/agents/src/voice/agent_session.ts:263:17)
              at MultiUserTranscriber.startSession (/Users/mac/Documents/side/transcriber-node/src/agent.ts:209:19)
    }
Can anyone confirm if this is a node specific issue? and multiple participants and multiple agents are not yet supported on node
r
@bitter-ghost-46918 does this workin in agents-js?