AI voices can Laugh, Sign and Whisper now!
In this short YouTube demo, John Bohan and three speakers showcase ElevenLabs' new V3 alpha AI voice model, highlighting its ability to produce realistic emotional cues such as laughter, whispering, sighing, and excitement through simple script tags (e.g., <whisper>, <excited>). The speakers demonstrate a whisper, and Bohan emphasizes how the technology can be controlled with these tags, making the AI output indistinguishable from human speech. The discussion then shifts to the broad implications for content creators: audiobook authors can voice entire novels with distinct character voices, podcasters and YouTubers can generate reliable co‑hosts, and language teachers can create native‑speaker lessons. Bohan notes that this single update could upend the roughly $3 billion voice‑acting industry. The overall tone is enthusiastic and forward‑looking, presenting the feature set as a game‑changing breakthrough.
