Heaptalk, Jakarta — ElevenLabs, an AI audio research and deployment startup headquartered in New York, US, introduced its latest ability to create conversational AI bots (11/18).
Developers can create conversational agents within the startup’s developer platform with diverse customization options, including tone of voice and response length. The platform allows users to determine the AI bot’s persona by choosing its primary language, first message, and system prompt.
In advance, users must decide which large language model to build, such as Gemini, GPT, or Claude. They can also choose the temperature of responses to determine how creative the response should be and the token usage limit. Tune aspects can be customized, for example, voice, latency, stability, authentication criteria, and maximum conversation length with the AI agent.
Sam Sklar, ElevenLabs’ Head of Growth, said many of its clients already use this ability to create conversational AI agents, as TechCrunch reported. However, the most challenging parts were integrating the knowledge base and handling customer interruptions. Hence, the startup built a complete pipeline for conversational bots.
In addition, users can add their knowledge base, including a file, URL, or text block, to power the conversational bot and integrate their own custom LLM with it. ElevenLabs’ SDK is compatible with Python, JavaScript, React, and Swift. The startup also offers a WebSocket API for more customization.
Companies can also define criteria for collecting specific data items—for instance, the name and email of customers speaking to the agent—along with evaluation criteria in natural language to define the success or failure of the call.
