ElevenLabs‘ new feature is a NotebookLM competitor for creating GenAI podcasts

Voice AI startup ElevenLabs on Wednesday launched a function that allows you to add various kinds of content material to create a multi-speaker podcast for you, much like Google’s NotebookLM.

The function, known as GenFM, may be discovered on the corporate’s ElevenLabs Reader iOS app. At the moment, GenFM helps 32 languages, together with English, Hindi, Portuguese, Chinese language, Spanish, French, German, Japanese, and others.

To make use of GenFM, you first add a YouTube video, textual content, or doc. The app then picks out two voices robotically to create a podcast. There are over a dozen voices obtainable, the corporate notes.

One of many attention-grabbing screens you would possibly see throughout the app making ready an AI-generated podcast is “Sprinkling some umms” and “including some considerate pauses.” At a time when tools are helping humans remove “Ums” and “Ahs,” ElevenLabs is including human components to its AI-generated podcast.

“We debated about how a lot to introduce ‘ums,’ ‘ahs,’ ‘mhmms’/ laughter/respiratory comparable human dialogue fillers or overlays — we’re aiming to strike the suitable steadiness of pure, human dialog and offering utility from the content material,” Jack McDermott, who takes care of cellular progress at ElevenLabs instructed TechCrunch over electronic mail.

“As we take a look at distinctive long-form podcasts on the market, we discover [fewer] interruptions and extra pure, insightful dialog movement,” he mentioned. “That kind of expertise is in the end what we’re aiming for whereas making audio narratives extra accessible throughout voices and languages.”