1. Home
  2. Classes
  3. EssentiaTFInputExtractor

new EssentiaTFInputExtractor( EssentiaWASM [, extractorType [, isDebug ] ] )

Parameters
Name Type Attributes Default Description
EssentiaWASM EssentiaWASM

Essentia WASM emcripten global module object
extractorType string <optional>
 'musicnn'

type of the desired extractor type (eg. 'muscinn', 'vggish' or 'tempocnn').
isDebug boolean <optional>
 false

whether to enable EssentiaWASM internal debugger for logs.
Details

machinelearning/tfjs_input_extractor.js, line 82

Members

essentia

Properties
Name Type Description
this.essentia EssentiaJS

an instance of EssentiaWASM.EssentiaJS.
this.extractorType string

type of the choosen extractor (eg. 'muscinn', 'vggish' or 'tempocnn').
Details

machinelearning/tfjs_input_extractor.js, line 89

Methods

arrayToVector( inputArray ) → {VectorFloat}

Description

Convert a typed JS Float32Array into VectorFloat type.

Parameters
Name Type Description
inputArray Float32Array

input Float32 typed array.
Returns

returns converted VectorFloat array.

Details

machinelearning/tfjs_input_extractor.js, line 112

vectorToArray( inputVector ) → {Float32Array}

Description

Convert an input VectorFloat array into typed JS Float32Array

Parameters
Name Type Description
inputVector VectorFloat

input VectorFloat array
Returns

returns converted JS typed array

Details

machinelearning/tfjs_input_extractor.js, line 122

<async> getAudioBufferFromURL( audioURL, webAudioCtx ) → {Promise.<AudioBuffer>}

Description

Decode and returns the audio buffer from an given audio url or blob uri using Web Audio API. (NOTE: This doesn't work on Safari browser)

Parameters
Name Type Description
audioURL string

web url or blob uri of a audio file
webAudioCtx AudioContext

an instance of Web Audio API AudioContext
Returns

decoded audio buffer as a promise

Details

machinelearning/tfjs_input_extractor.js, line 134

audioBufferToMonoSignal( buffer ) → {Float32Array}

Description

Convert an AudioBuffer object to a Mono audio signal array. The audio signal is downmixed to mono using essentia MonoMixer algorithm if the audio buffer has 2 channels of audio. Throws an expection if the input AudioBuffer object has more than 2 channels of audio.

Parameters
Name Type Description
buffer AudioBuffer

AudioBuffer object decoded from an audio file.
Returns

audio channel data. (downmixed to mono if its stereo signal).

Details

machinelearning/tfjs_input_extractor.js, line 162

downsampleAudioBuffer( sourceBuffer ) → {Float32Array}

Description

Downsample a audio buffer to a target audio sample rate using the Web Audio API NOTE: This method will only works on web-browsers which supports the Web Audio API.

Parameters
Name Type Description
sourceBuffer AudioBuffer

AudioBuffer object decoded from an audio file.
Returns

decoded audio buffer object

Details

machinelearning/tfjs_input_extractor.js, line 182

compute( audioFrame ) → {EssentiaTFInputExtractorOutput}

Description

This method compute the pre-configured features for a given audio signal frame. It throws an exception if the size of audioFrame is not equal to the pre-configured audioFrame size for the selected extractorType setting.

Parameters
Name Type Description
audioFrame Float32Array

a frame of audio signal as Float32 typed JS array.
Returns

returns the computed feature for the input the given audio frame.

Details

machinelearning/tfjs_input_extractor.js, line 212

computeFrameWise( audioSignal, hopSize? ) → {EssentiaTFInputExtractorOutput}

Description

This method compute the pre-configured feature for a whole audio signal. It is a wrapper on top of the compute method. It throws an exception if the size of audioFrame is not equal to the pre-configured size.

Parameters
Name Type Description
audioSignal Float32Array

decoded audio signal as Float32 typed JS array.
hopSize? number

optional param for specifying hopSize for overlapping-frames. By default use none.
Returns

returns the computed frame-wise feature for the given audio signal.

Details

machinelearning/tfjs_input_extractor.js, line 270

"delete"() → {null}

Description

Delete essentia session and frees the memory.

Returns
Details

machinelearning/tfjs_input_extractor.js, line 295

shutdown() → {null}

Description

This method shutdown all the instance of Essentia WASM and frees the memory. NOTE: If you want to just free the memory of the pre-configured extractor, use this.extractor.delete() instead.

Returns
Details

machinelearning/tfjs_input_extractor.js, line 306