---
canonical_url: "https://docs.composio.dev/toolkits/audd"
markdown_url: "https://docs.composio.dev/toolkits/audd.md"
product: "platform"
version: null
last_updated: null
---

# AudD

AudD identifies music from audio and supports lyrics search, long-file recognition, stream monitoring, and private music catalogs.

- **Category:** video & audio
- **Auth:** API_KEY
- **Composio-managed OAuth available?** N/A
- **Tools:** 11
- **Triggers:** 0
- **Slug:** `AUDD`
- **Version:** 00000000_00

## Tools

### Add Stream

**Slug:** `AUDD_ADD_STREAM`

Start persistent recognition for a live audio, radio, Twitch, or YouTube stream; this can consume a paid stream slot billed monthly until deleted.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `radio_id` | integer | Yes | Caller-chosen stream identifier that must be unique within the connected AudD account and between 0 and 2147483647. |
| `stream_url` | string | Yes | Direct DASH, Icecast, HLS, m3u, or m3u8 URL, or a twitch:, youtube:, or youtube-ch: shortcut. |
| `callback_timing` | string ("after" | "before") | No | Send callbacks after each song ends (default, includes play_length) or before the song starts. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Delete Stream

**Slug:** `AUDD_DELETE_STREAM`

Stop recognition and permanently delete one monitored stream, releasing its paid stream slot and stopping its ongoing monthly billing.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `radio_id` | integer | Yes | ID of the monitored stream to stop and permanently delete (0-2147483647). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Stream Callback URL

**Slug:** `AUDD_GET_STREAM_CALLBACK_URL`

Return the account-wide callback URL currently configured for AudD stream-recognition events.

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Streams

**Slug:** `AUDD_LIST_STREAMS`

List every monitored stream configured on the connected AudD account, including its ID, source URL, running state, and long-poll category.

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Recognize Audio

**Slug:** `AUDD_RECOGNIZE_AUDIO`

Identify one song from the first 12 seconds of an audio URL, uploaded file, or base64 audio; returns null when no song matches and consumes recognition quota.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file` | object | No | Audio file up to 10 MB; provide exactly one of file, source_url, or audio_base64. |
| `market` | string | No | Country or storefront code used for regional metadata results. |
| `source_url` | string | No | Public HTTPS URL of the audio file; provide exactly one of source_url, file, or audio_base64. |
| `audio_base64` | string | No | Base64-encoded audio for clients that cannot provide a URL or file; this provider option is minimally supported. |
| `metadata_providers` | string | No | Comma-separated extra metadata providers: apple_music, spotify, deezer, or musicbrainz. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Recognize Long Media

**Slug:** `AUDD_RECOGNIZE_LONG_MEDIA`

Scan selected 12-second chunks of a long audio or video URL or file for matching songs on AudD's enterprise host; file uploads are capped at 20 MB, and each scanned chunk is billable after the free allowance, so limit defaults to 1.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file` | object | No | Audio or video file up to 20 MB. Provide exactly one of file or source_url; use source_url for larger media. |
| `skip` | integer | No | Number of 12-second chunks to skip after each scanned group. |
| `every` | integer | No | Number of consecutive 12-second chunks to scan before applying skip. |
| `limit` | integer | No | Maximum billable 12-second chunks to scan. Defaults to 1 as a cost guard; raise deliberately for broader scans. |
| `source_url` | string | No | Public URL of the long audio or video. Provide exactly one of source_url or file. |
| `use_timecode` | boolean | No | Use t, time_continue, or start from the source URL. Cannot be true with skip_first_seconds. |
| `response_type` | string ("json" | "csv") | No | Return structured JSON matches or CSV text. |
| `accurate_offsets` | boolean | No | Whether to calculate more accurate start and end offsets for matches. |
| `metadata_providers` | string | No | Comma-separated extra metadata providers. Spotify is not supported by this enterprise endpoint. |
| `skip_first_seconds` | integer | No | Seconds to skip at the start. Cannot be combined with use_timecode=true. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Search Lyrics

**Slug:** `AUDD_SEARCH_LYRICS`

Search AudD's lyrics index by artist, title, or lyric text and return up to the provider's default 10 matches with full lyrics when available.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `query` | string | Yes | Lyrics search text, such as an artist and song title or a remembered lyric phrase. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Set Stream Callback URL

**Slug:** `AUDD_SET_STREAM_CALLBACK_URL`

Replace the account-wide URL that receives AudD stream-recognition callbacks; this persistently affects every monitored stream.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `callback_url` | string | Yes | HTTPS callback receiver URL. Use https://audd.tech/empty/ only when intentionally using long polling without delivered callbacks. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Update Stream URL

**Slug:** `AUDD_UPDATE_STREAM_URL`

Replace the source URL of an existing monitored stream while keeping its account-specific radio ID.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `radio_id` | integer | Yes | ID of the existing monitored stream to update (0-2147483647). |
| `stream_url` | string | Yes | Replacement DASH, Icecast, HLS, m3u, m3u8, Twitch, or YouTube stream source. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Upsert Custom Catalog Song

**Slug:** `AUDD_UPSERT_CUSTOM_CATALOG_SONG`

Fingerprint a URL or uploaded song into the private catalog under a caller-chosen audio ID; reusing the ID replaces its fingerprint and requires provider-enabled paid access.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file` | object | No | Song file to fingerprint; provide exactly one of file or source_url. |
| `audio_id` | integer | Yes | Caller-chosen private-catalog ID returned by later matches; reusing an ID replaces that catalog slot's fingerprint. |
| `source_url` | string | No | Public URL of the song to fingerprint; provide exactly one of source_url or file. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Wait for Stream Events

**Slug:** `AUDD_WAIT_FOR_STREAM_EVENTS`

Hold one request for up to 90 seconds waiting for recognition or notification events from a stream long-poll category; use an available event time as since_time on the next call to resume without duplicates.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `timeout` | integer | No | Seconds the provider should hold this request while waiting for an event. |
| `since_time` | integer | No | Timestamp in milliseconds from the previous response, used to resume and deduplicate events. |
| `longpoll_category` | string | Yes | Nine-character longpoll_category returned by List Streams for the target stream. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |
