Camb.ai icon

Camb.ai

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

Actions9

Audio → Generate Sound

AI-generated

Summary

The operation generates sound or music from a text prompt using the Camb.ai text-to-sound service. It requires a prompt, audio type (sound or music), and a duration between 1 and 10 seconds (default 8 seconds). The workflow assigns a task ID, polls for completion, and returns the generated audio as a binary data object linked to a specified output field. No text input, voice cloning, or separation of existing audio is involved.

Inputs

  • prompt (required) — Description of the sound to generate
  • audioType — Type of audio to generate (default: sound)
  • duration — Duration of generated audio in seconds (default 8, max 10)
  • outputFormat — Audio output format (default wav)
  • binaryPropertyName (required) — Name of the output binary field to contain the audio file

Output shape

A JSON object containing taskId, runId, prompt, audioType, duration, and size; plus a binary field named by binaryPropertyName containing the audio file.

The operation returns immediately with a task ID. The workflow polls for completion, then retrieves the audio file. The returned JSON includes task and run identifiers plus the generated audio metadata. The binary payload is placed in the field named by binaryPropertyName.

Examples

Example 1: A workflow author sets prompt='ocean waves crashing', audioType=music, duration=5, outputFormat=wav, binaryPropertyName='audioClip'. The node returns a task ID, polls until the sound file is ready, and places a WAV-encoded audio clip named 'sound_{taskId}.wav' into the 'audioClip' field of the workflow item.

The author provides a prompt describing the desired sound, selects the type (sound or music), sets the duration (1-10 seconds), chooses an output format, and names the field that will receive the resulting audio file.

Links

Discussion