Camb.ai icon

Camb.ai

Use Camb.ai for TTS, transcription, translation, voice cloning, and more

Actions9

Audio → Separate Audio

AI-generated

Summary

Separates foreground speech from background audio using Camb.ai, returning both as binary outputs along with task completion metadata.

Inputs

  • inputBinaryField (required) — Name of the binary field containing the audio file to separate.
  • binaryPropertyName (required) — The name of the output binary field to contain the separated foreground audio.

Output shape

binary + json

The node reads the input audio binary, sends it to Camb.ai's /audio-separation endpoint, polls for completion, then downloads and returns both foreground (speech) and background audio as WAV binary buffers. JSON output includes task and run IDs plus URLs for both audio streams.

Examples

Example 1: User provides an audio clip in the `data` binary field and sets `binaryPropertyName` to `audio`.

The node outputs two binary fields: audio containing the extracted foreground speech, and audio_background containing the residual background audio. JSON includes taskId, runId, and URLs for both streams.

Links

Discussion