If there’s one area where AI technology is really thriving, it’s in converting text into realistic-sounding speech. Players like NaturalReader and Speechify are using AI to narrate text using a variety of highly authentic-sounding voices, including those of some celebrities.

But now there’s an even stronger contender: ElevenLabs, an AI audio startup that’s not only leveraging existing technologies, but pushing them to new heights. The voices it offers are simply incredible. Everything from pitch, timbre, pauses, pronunciations, intonations, fluidity, context awareness—it’s all too good to be true. Forget Gemini Live or ChatGPT’s Voice Mode. This is what I’ve been crying with shameless joy over for a while now.

ElevenLabs is an AI-based company that develops natural-sounding speech, sound effects, and voices. The company has multiple products and various features that power them, including text to speech, speech to speech, text to SFX, voice cloning, and voice isolation.

The startup recently released its Reader app , which is where the real game now lies, as it’s completely free to use (I checked) and only requires a subscription if you use it on the web. With support for 32 languages and dozens of professional voices, including those of a few deceased celebrities, as well as the ability to create your own voice clone, ElevenLabs has quickly become the go-to service for publishing, podcasts, audiobooks, and more.