process#
audioclass.batch.process
#
Module for batch processing audio data with classification models.
Attributes#
Classes#
Functions#
process_iterable(process_array, iterable, tags, name, confidence_threshold=DEFAULT_THRESHOLD)
#
Process an iterable of audio data using a models process_array method.
This function iterates over batches of audio clips, processes each clip
using the provided process_array function, and returns a list of
ClipPrediction objects. The process_array function should take a
numpy array of audio data and return a tuple of class probabilities and
extracted features.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
process_array |
Callable[[ndarray], Tuple[ndarray, ndarray]]
|
A function that takes a numpy array of audio data and returns a tuple of class probabilities and extracted features. |
required |
iterable |
BaseIterator
|
An iterator that yields batches of audio clips. |
required |
tags |
List[Tag]
|
A list of tags that the model can predict. |
required |
name |
str
|
The name of the model. |
required |
confidence_threshold |
float
|
The minimum confidence threshold for a tag to be included in a
prediction. Defaults to |
DEFAULT_THRESHOLD
|
Returns:
| Type | Description |
|---|---|
List[ClipPrediction]
|
A list of |