Recipes
Find end-to-end HUMAIN Voice recipes for speaker-labelled transcripts, live captions, and speech files.
Each recipe is one complete task in JavaScript and Python. Install the SDK first with the Quickstart; to pick a workflow, see Concepts.
Transcribe a recording with speaker labels
Turn a long recording into speaker turns and WebVTT captions.
Realtime transcription
Show live captions that settle into a final transcript.
Text to speech: save or stream
Save speech to a WAV file or play it as it arrives.
Emotion and speaking style
Make a voice sound happy, whisper, or read like a newscaster.
Voice a support agent
Make an apology sound sorry and a goodbye sound warm.
Narrate a story
Read a story aloud with whispered and fearful moments.
Let an LLM write tagged text
Prompt a voice agent's LLM to add tags, then check them.
Talk to a voice agent from a web page
Start a Voice Agent session from your backend and join it from the browser.
Voice Agent browser SDK
Join a Voice Agent call from the browser with @humain-voice/agent, handle client tools, and react to status, transcripts, warnings, and end reasons.
Transcribe a recording with speaker labels
Turn a long recording into a speaker-labelled transcript and WebVTT captions with a batch job.