Configuration Reference

This page documents every option you can pass to the POST /v2/edits endpoint (and the SDK process() / create_edit() methods).

Input payload

Single file

{
  "input": {
    "files": ["https://example.com/episode.mp3"]
  }
}

Multiple files (multi-track)

Pass an array to process multi-speaker recordings with separate audio tracks:

{
  "input": {
    "files": [
      "https://example.com/host.mp3",
      "https://example.com/guest.mp3"
    ],
    "upload_type": "multitrack"
  }
}

Audio cleaning

These options detect and remove unwanted sounds from the recording.

fillers

Remove filler words such as "um", "uh", "like", "you know", and similar hesitation markers.

long_silences

Trim long pauses and awkward gaps between sentences.

mouth_sounds

Remove clicks, lip smacks, tongue sounds, and similar mouth noises that often appear between words.

breath

Remove audible breathing sounds between sentences and during pauses.

breath options:

stutters

Detect and remove repeated word fragments (e.g. "I— I— I think").

Audio enhancement

These options improve audio quality without cutting content.

{
  "remove_noise": true,
  "normalize": true
}

remove_noise

Reduce background noise: hiss, hum, fan noise, AC, street noise, and similar sounds.

true is Noise v2 (recommended). Omitting the key on /v2/edits and /v2/noise is the same as true. "legacy" pins the previous engine (DeepFilterNet). "v2" is still accepted and means the same as true. false turns noise removal off.

remove_noise values:

studio_sound

Apply aggressive audio enhancement to make the recording sound studio-quality. Best used on voice-only recordings in quiet environments.

studio_sound options:

normalize

Normalize v2. Levels loudness across the recording so it stays consistent. This option is a boolean: true runs the current leveler (Normalize v2). There is no "v2" string.

Turn it on together with remove_noise: true. When normalize is true and target_lufs is omitted, loudness targets -16 LUFS. Pass target_lufs to override that. Accepted values are -36 to -6.

autoeq

Legacy automatic EQ correction. Prefer studio_sound; autoeq will be removed in a future release.

mute_lufs + target_lufs

target_lufs sets the integrated loudness used when normalize is true. -16 is the podcast target and the Normalize v2 default when the field is omitted. Accepted range: -36 to -6. mute_lufs is the gate used while measuring loudness.

{
  "config": {
    "mute_lufs": -120,
    "target_lufs": -16
  }
}

Output format

export_format

The audio format for the cleaned output file. This only applies to audio outputs. Video jobs keep the original video container format.

Content generation

transcription

Return a full word-by-word transcript of the audio.

Language is auto-detected. See supported languages.

summarize

Generate chapter markers, key learnings, and an episode summary from the transcript. Automatically enables transcription.

social_content

Generate social media post suggestions (tweets, LinkedIn, show notes) from the content.

Additional options

These options are also supported by the current SDK/backend surface but are not covered in the sections above.

Additional cleanup and enhancement

Additional output and delivery

Additional advanced workflow options

Full example

result = client.process(
    "https://example.com/episode.mp3",
    fillers=True,
    long_silences=True,
    mouth_sounds=True,
    breath=True,
    stutters=True,
    remove_noise=True,
    studio_sound=False,
    normalize=True,
    mute_lufs=-120,
    target_lufs=-16,
    export_format="mp3",
    transcription=True,
    summarize=True,
    social_content=False,
)

const result = await client.process(
  'https://example.com/episode.mp3',
  {
    fillers: true,
    long_silences: true,
    mouth_sounds: true,
    breath: true,
    stutters: true,
    remove_noise: true,
    studio_sound: false,
    normalize: true,
    mute_lufs: -120,
    target_lufs: -16,
    export_format: 'mp3',
    transcription: true,
    summarize: true,
    social_content: false,
  }
);

{
  "input": {
    "files": ["https://example.com/episode.mp3"],
    "config": {
      "fillers": true,
      "long_silences": true,
      "mouth_sounds": true,
      "breath": true,
      "stutters": true,
      "remove_noise": true,
      "studio_sound": false,
      "normalize": true,
      "mute_lufs": -120,
      "target_lufs": -16,
      "export_format": "mp3",
      "transcription": true,
      "summarize": true,
      "social_content": false
    }
  }
}

Recommended presets