Higgs VoiceOver puts AI voice-over inside your video editor: direct each read for emotion, style and speed, generate word-timed subtitles, and build a custom voice from your own audio. Powered by Higgs TTS 3, available today for DaVinci Resolve Studio and Adobe Premiere Pro.

The Higgs TTS 3 API now rewrites input text the way a person would read it aloud, expanding numbers, dates, currency, URLs and abbreviations into spoken form. It can also return the start and end time of every word alongside the audio, so text can stay in step with speech.

pipecat-boson brings Higgs Realtime into Pipecat as a Realtime LLM service. One model takes live audio in and streams audio back, replacing the speech-to-text, language model, and text-to-speech chain — with interruption handling, function calling, 100+ languages, and audio out at $0.014 a minute.

Real-time voice AI is easy to demo and much harder to build well. The Higgs Realtime API Tutorial is a build-it-yourself guide that constructs a browser voice assistant one layer at a time — live audio, interruptions, out-of-order events, tool calling — with a Git checkpoint and acceptance test at every stage.

A real-time voice model and API that handles interruptions, adapts mid-sentence, and carries conversations the way people actually talk. It leads major deployed real-time APIs on audio benchmarks, supports 100+ languages, and is compatible with the OpenAI Realtime API.

A benchmark for post-interruption recovery in voice agents — whether what the model says after you cut in resumes the workflow correctly. 428 interruption samples, 6 types, 10 enterprise domains, closed- and open-weight models.