Recipes Overview
Choose an end-to-end HUMAIN Voice task by its input, output, and completion criterion.
Recipes begin after setup and end with a usable artifact or finalized state. Choose by what your app has as input—not by which transport sounds familiar.
Before you begin
- Use JavaScript or Python SDK
0.18.0and complete the Quickstart unless you already made a tested request. - Have the current
API_URLandAPI_KEY; Socket.IO defaults to/socket.io. Keep the key in a trusted server environment. - Use representative audio or text and choose a destination for the completed artifact before running the recipe.
CI compiles or type-checks every displayed JavaScript and Python program from
the same fixture against SDK 0.18.0.
Choose by input
Complete long-form media → speaker-labelled transcript
Input: a meeting, podcast, archive, or other complete recording. Output: transcript, finalized WebVTT, and speaker attribution. Done: the batch job succeeds, word timing is reconciled, and output files are written.
Arriving audio → finalized captions
Input: PCM16 audio while it is still arriving. Output: replaceable provisional text and finalized WebVTT. Done: final state is recorded, captions are written, and the client is disconnected.
Text → playable WAV
Input: text and a runtime voice. Output: PCM16 wrapped in a valid WAV container. Done: the final audio chunk is received, the WAV file is written, and the client is closed.
Where fast transcription fits
Fast transcription is a separate SDK path for a complete audio unit whose latency matters, such as one conversational turn. It sends the whole encoded unit and then receives result events. It's not live microphone streaming, and this documentation doesn't route meetings, podcasts, archives, or other long-form media through it—use the batch recipe for those.
Start from the SDK guide when fast transcription matches the input.
Next steps
A completed recipe establishes its stated artifact and cleanup path. Before launch, repeat it with representative inputs and exercise deadlines, terminal states, disconnects, retries, capacity backpressure, empty voice lists, and secret handling.