Vermillio Music SDK Reference¶

Vermillio Music AI Detect Client¶

The VermillioMusicAIDetect client is the primary interface for interacting with the Music AI Detect Service.

Bases: VermillioBasePipelineClient

init ¶

__init__(config: Optional[VermillioConfig] = None)

Initializes the Music AI Detection Client.

Parameters:

Name	Type	Description	Default
`config`	`Optional[VermillioConfig]`	An optional configuration object. If None, it attempts to load configuration from environment variables.	`None`

load ¶

load(
    sources: Union[list[AIDetectExternalSource], list[str]],
    context: Optional[str] = None,
) -> Optional[list[AIDetectPipelineSource]]

Loads a list of audio sources into the AI detection pipeline. Processing occurs asynchronously.

Parameters:

Name	Type	Description	Default
`sources`	`Union[list[AIDetectExternalSource], list[str]]`	The external audio sources, or paths to a file, to analyze.	required
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`

Returns:

Type	Description
`Optional[list[AIDetectPipelineSource]]`	Optional[AIDetectPipelineSource]: An AI Detect Pipeline Source entry, if was loaded successfully.

load_results ¶

load_results(
    sources: Union[list[AIDetectExternalSource], list[str]],
    context: Optional[str] = None,
    wait: bool = True,
) -> Optional[list[AIDetectPipelineResults]]

Loads a list of audio sources into the AI detection pipeline and returns results.

Parameters:

Name	Type	Description	Default
`sources`	`Union[list[AIDetectExternalSource], list[str]]`	The external audio sources, or paths to a file, to analyze.	required
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`
`wait`	`bool`	Whether to wait until each source has finished processing before returning.	`True`

Returns:

Type	Description
`Optional[list[AIDetectPipelineResults]]`	Optional[list[AIDetectPipelineResults]]: Detection results for each loaded source.

pipeline ¶

pipeline() -> AIDetectPipeline

Retrieves the configuration or definition of the current AI detection pipeline.

Returns:

Name	Type	Description
`AIDetectPipeline`	`AIDetectPipeline`	An object containing the pipeline definition.

results ¶

results(
    id_or_source_id: str,
    context: Optional[str] = None,
    wait: bool = False,
) -> Optional[AIDetectPipelineResults]

Retrieves the AI detection analysis results for a specific source.

Parameters:

Name	Type	Description	Default
`id_or_source_id`	`str`	The unique identifier of the source to get results for.	required
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`
`wait`	`bool`	Optionally wait until the source is in a done state by retrying on a response indicating incomplete.	`False`
`timeout`	`int`		required

Returns:

Type	Description
`Optional[AIDetectPipelineResults]`	Optional[AIDetectPipelineResults]: An object containing the AI vs. Real probability scores and analysis metadata.

run ¶

run(
    source: Union[AIDetectExternalSource, str],
    context: Optional[str] = None,
    timeout: float = 60,
    poll_rate: float = 1,
) -> Optional[AIDetectPipelineSource]

Loads a source and waits (within a timeout) for the AI detection analysis to complete.

Parameters:

Name	Type	Description	Default
`source`	`Union[AIDetectExternalSource, str]`	The external audio source, or path to a file, to analyze.	required
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`
`timeout`	`float`	Optional maximum time in seconds to wait for the analysis.	`60`
`poll_rate`	`float`	Optional frequency in seconds at which to poll for updates.	`1`

Returns:

Type	Description
`Optional[AIDetectPipelineSource]`	Optional[list[AIDetectPipelineSource]]: The list of processed sources upon completion.

run_results ¶

run_results(
    source: Union[AIDetectExternalSource, str],
    context: Optional[str] = None,
    timeout: float = 60,
    poll_rate: float = 1,
) -> Optional[AIDetectPipelineResults]

Loads a source and waits for completion, returning detailed AI detection analysis results.

Parameters:

Name	Type	Description	Default
`source`	`Union[AIDetectExternalSource, str]`	The external audio source, or path to a file, to analyze.	required
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`
`timeout`	`float`	Maximum time in seconds to wait for results.	`60`
`poll_rate`	`float`	Frequency in seconds at which to poll for results.	`1`

Returns:

Type	Description
`Optional[AIDetectPipelineResults]`	Optional[AIDetectPipelineResults]: A list containing detection scores, confidence levels, and AI-generated probability markers.

source ¶

source(
    id_or_source_id: str, context: Optional[str] = None
) -> Optional[AIDetectPipelineSource]

Fetches details for a specific audio source.

Parameters:

Name	Type	Description	Default
`id_or_source_id`	`str`	The unique identifier or source ID to retrieve.	required
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`

Returns:

Type	Description
`Optional[AIDetectPipelineSource]`	Optional[AIDetectPipelineSource]: The source object if found.

sources ¶

sources(
    context: Optional[str] = None,
    after_id: Optional[str] = None,
    limit: int = 10,
) -> Optional[list[AIDetectPipelineSource]]

Lists previously loaded audio sources with pagination support.

Parameters:

Name	Type	Description	Default
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`
`after_id`	`Optional[str]`	The optional ID to start listing after for pagination.	`None`
`limit`	`int`	The maximum number of sources to return.	`10`

Returns:

Type	Description
`Optional[list[AIDetectPipelineSource]]`	Optional[list[AIDetectPipelineSource]]: A list of source objects if found.

status ¶

status(
    context: Optional[str] = None,
) -> AIDetectPipelineStatus

Checks the health or operational status of the AI Detection service.

Parameters:

Name	Type	Description	Default
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`

Returns:

Name	Type	Description
`AIDetectPipelineStatus`	`AIDetectPipelineStatus`	An object indicating the current state of the pipeline.

upload_and_load ¶

upload_and_load(
    file_paths: list[str],
    titles: list[str],
    context: Optional[str] = None,
) -> Optional[list[AIDetectPipelineSource]]

Uploads audio files as assets and loads them into the AI detection pipeline. Processing occurs asynchronously.

Parameters:

Name	Type	Description	Default
`file_paths`	`list[str]`	Local file paths of audio files to upload.	required
`titles`	`list[str]`	Titles for each uploaded asset.	required
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`

Returns:

Type	Description
`Optional[list[AIDetectPipelineSource]]`	Optional[list[AIDetectPipelineSource]]: The loaded pipeline source entries.

upload_and_load_results ¶

upload_and_load_results(
    file_paths: list[str],
    titles: list[str],
    context: Optional[str] = None,
    wait: bool = True,
) -> Optional[list[AIDetectPipelineResults]]

Uploads audio files as assets, loads them into the AI detection pipeline, and returns results.

Parameters:

Name	Type	Description	Default
`file_paths`	`list[str]`	Local file paths of audio files to upload.	required
`titles`	`list[str]`	Titles for each uploaded asset.	required
`context`	`Optional[str]`	An optional context, uses `AIDetectPipeline.default_context` if missing.	`None`
`wait`	`bool`	Whether to wait until each source has finished processing before returning.	`True`

Returns:

Type	Description
`Optional[list[AIDetectPipelineResults]]`	Optional[list[AIDetectPipelineResults]]: Detection results for each loaded source.

Music AI Detect Model Classes¶

These Pydantic models represent the inputs and outputs of the Music AI Detect Service.

AIDetectExternalSource `pydantic-model` ¶

Bases: ExternalSource

Fields:

Validators:

require_source

path `pydantic-field` ¶

path: Optional[str] = None

An accessible url to download the track from, we support http(s)://, public gs://, public s3://, etc.

AIDetectLoadRequest `pydantic-model` ¶

Bases: PipelineLoadRequest

Fields:

context (Optional[str])
sources (Union[list[AIDetectExternalSource], list[str]])

sources `pydantic-field` ¶

sources: Union[list[AIDetectExternalSource], list[str]]

List of external sources (or paths) to run AI Detection on.

AIDetectPipelineResults `pydantic-model` ¶

Bases: PipelineResults

Fields:

id (str)
status (PipelineSourceStatus)
created_at (float)
updated_at (Optional[float])
source_id (Optional[str])
source_path (Optional[str])
results (Optional[AIDetectResults])

results `pydantic-field` ¶

results: Optional[AIDetectResults] = None

The AI Detection results, if the execution was successful.

AIDetectPipelineSource `pydantic-model` ¶

Bases: PipelineSource

Fields:

id (str)
status (PipelineSourceStatus)
created_at (float)
updated_at (Optional[float])
source_id (Optional[str])
media_id (Optional[str])
asset_id (Optional[str])
path (Optional[str])

path `pydantic-field` ¶

path: Optional[str]

An accessible url to download the track from, we support http(s)://, public gs://, public s3://, etc.

AIDetectResults `pydantic-model` ¶

Bases: BaseCamelModel

Fields:

detections (list[AIDetectSegment])

detections `pydantic-field` ¶

detections: list[AIDetectSegment] = []

List of detection segments

AIDetectRunRequest `pydantic-model` ¶

Bases: PipelineRunRequest

Fields:

context (Optional[str])
source (Union[AIDetectExternalSource, str])

source `pydantic-field` ¶

source: Union[AIDetectExternalSource, str]

The external source, or path of the fule to run AI Detection on.

AIDetectSegment `pydantic-model` ¶

Bases: BaseCamelModel

Fields:

query_segment (AudioSegment)
label (Literal['real_music', 'ai_music'])
confidence (Optional[float])
source_id (str)

confidence `pydantic-field` ¶

confidence: Optional[float] = None

Confidence in the result.

label `pydantic-field` ¶

label: Literal['real_music', 'ai_music']

Whether or not this segment is real or ai.

query_segment `pydantic-field` ¶

query_segment: AudioSegment

The times in the original source file that this result is associated with.

source_id `pydantic-field` ¶

source_id: str

The source (if available) of labelled 'ai_music', if unknown will be 'ai_music'. 'real_music' if labelled as real music.

AudioSegment `pydantic-model` ¶

Bases: BaseCamelModel

Fields:

start (float)
end (float)

end `pydantic-field` ¶

end: float

The end time (inclusive) in seconds of the segment.

start `pydantic-field` ¶

start: float

The start time (inclusive) in seconds of the segment.

Vermillio Music SDK Reference¶

Vermillio Music AI Detect Client¶

__init__ ¶

load ¶

load_results ¶

pipeline ¶

results ¶

run ¶

run_results ¶

source ¶

sources ¶

status ¶

upload_and_load ¶

upload_and_load_results ¶

Music AI Detect Model Classes¶

AIDetectExternalSource pydantic-model ¶

path pydantic-field ¶

AIDetectLoadRequest pydantic-model ¶

sources pydantic-field ¶

AIDetectPipelineResults pydantic-model ¶

results pydantic-field ¶

AIDetectPipelineSource pydantic-model ¶

path pydantic-field ¶

AIDetectResults pydantic-model ¶

detections pydantic-field ¶

AIDetectRunRequest pydantic-model ¶

source pydantic-field ¶

AIDetectSegment pydantic-model ¶

confidence pydantic-field ¶

label pydantic-field ¶

query_segment pydantic-field ¶

source_id pydantic-field ¶

AudioSegment pydantic-model ¶

end pydantic-field ¶

start pydantic-field ¶

init ¶

AIDetectExternalSource `pydantic-model` ¶

path `pydantic-field` ¶

AIDetectLoadRequest `pydantic-model` ¶

sources `pydantic-field` ¶

AIDetectPipelineResults `pydantic-model` ¶

results `pydantic-field` ¶

AIDetectPipelineSource `pydantic-model` ¶

path `pydantic-field` ¶

AIDetectResults `pydantic-model` ¶

detections `pydantic-field` ¶

AIDetectRunRequest `pydantic-model` ¶

source `pydantic-field` ¶

AIDetectSegment `pydantic-model` ¶

confidence `pydantic-field` ¶

label `pydantic-field` ¶

query_segment `pydantic-field` ¶

source_id `pydantic-field` ¶

AudioSegment `pydantic-model` ¶

end `pydantic-field` ¶

start `pydantic-field` ¶