Dataset Management and Labeling

Ingest, create, and label large data sets

Use the audioDatastore object to access data and perform common management tasks such as splitting data into train, validation, and test sets and distributing computations among workers. To label existing data sets or to create new data sets, use the Signal Labeler app.

Apps

Signal Labeler

Label signal attributes, regions, and points of interest, and extract features

Objects

audioDatastore Datastore for collection of audio files

Functions

`detectSpeech`	Detect boundaries of speech in audio signal (Since R2020a)
`detectspeechnn`	Detect boundaries of speech in audio signal using AI (Since R2023a)
`classifySound`	Classify sounds in audio signal (Since R2020b)
`speech2text`	Transcribe speech signal to text (Since R2022b)

Topics

Use Datastores to Manage Audio Data Sets
Use datastores to manage different data sets containing audio data and labels.
Choose an App to Label Ground Truth Data
Decide which app to use to label ground truth data: Image Labeler, Video Labeler, Ground Truth Labeler, Lidar Labeler, Signal Labeler, or Medical Image Labeler.
Label Spoken Words in Audio Signals
Use Signal Labeler to label spoken words in an audio signal.

Featured Examples

Import Audacity Labels to Signal Labeler

Import labels created in Audacity™ into Signal Labeler.

Open Live Script

Train Spoken Digit Recognition Network Using Out-of-Memory Audio Data

Trains a spoken digit recognition network on out-of-memory audio data using a transformed datastore. In this example, you apply a random pitch shift to audio data used to train a convolutional neural network (CNN). For each training iteration, the audio data is augmented using the audioDataAugmenter object and then features are extracted using the audioFeatureExtractor object. The workflow in this example applies to any random data augmentation used in a training loop. The workflow also applies when the underlying audio data set or training features do not fit in memory.

Open Live Script