API ReferenceSearch Metadata Filter

Audio Sound (with Metadata Filters)

Triggers an asynchronous search for video segments containing the specified `audio_class` (e.g., 'Music', 'Dog', 'Explosion') within the specified dataset (`dataset_id`), restricted to videos whose metadata matches every entry in `metadata_filters` (e.g., `genre = documentary`). Audio chunks are scored using the Audio Spectrogram Transformer (AST) model, which predicts confidence scores for 527 AudioSet sound classes across fixed-length audio segments extracted from ingested videos; no text embedding is computed. Returns the Databricks `run_id` immediately; use the status and result endpoints to poll for the final ranked list of sound segments (each with the parent video, time boundaries, the nearest keyframe, score, and optional moderation score).

Authentication

AuthorizationBearer

Bearer authentication of the form Bearer <token>, where token is your auth token.

Request

This endpoint expects an object.
dataset_idstringRequiredformat: "uuid"
Dataset to search within
audio_classenumRequired

AudioSet sound class label to search for (e.g. AudioClass.DOG, AudioClass.MUSIC)

limitintegerOptional1-100Defaults to 40
Max number of results to return
offsetintegerOptional>=0Defaults to 0
Number of results to skip before returning
metadata_filterslist of objectsOptional
List of filter objects applied against asset metadata fields before scoring. All filters are combined with AND.
skip_moderationbooleanOptionalDefaults to false

When true, moderation scoring is skipped and moderation_score will be null on all results.

Response

Successful Response
run_idinteger
Databricks job run ID to use for status polling and result retrieval

Errors

422
Unprocessable Entity Error