Configure the audio format
- Open your agent in the ElevenLabs agent dashboard.
- In the Advanced tab, set the user input audio format to PCM 16000Hz and save the change. Anam also supports other PCM sample rates and resamples audio automatically.
- Keep the agent ID available. You will use it when connecting the browser in step three.
Recommended settings
Adjust these settings in the same dashboard to suit your application. LLM choice. Choose a low-latency model and test its response time with your agent’s prompts and tools. If the model supports reasoning, start with a low reasoning budget or effort level. See ElevenLabs’ model selection guide for current options. If slower LLM responses are a problem, set a soft timeout in the Advanced tab (e.g. 2 seconds) and configure it to either use a fixed filler phrase or generate one with a faster LLM while the main response generates. Voices. In the Agent Voice settings, select Eleven v4 Turbo (eleven_v4_turbo). We recommend it over Eleven v4 (eleven_v4) for higher throughput and lower latency. Both models support audio tags, and existing v3 Conversational agents can keep using the same tags. Anam maps recognized tags, such as [laughs], [whispers], and [gasps], to Director Notes for the avatar’s performance. The model is configured in ElevenLabs, so changing it does not require changes to the Anam session code.
Background noise. If background noise is being picked up, turn on the Filter Background Speech toggle in the Advanced menu. It’s unclear whether this has an effect on latency.
Response speed. For the quickest responses, set Eagerness to Eager in the Advanced menu.
Your agent is configured. Next, connect it to Anam.
Step 2: Set up the server
Create the API route that fetches an ElevenLabs signed URL and returns an Anam session token.