Value Members

1.  final def !=(arg0: Any): Boolean

   Definition Classes

   AnyRef → Any

2.  final def ##(): Int

   Definition Classes

   AnyRef → Any

3.  final def ==(arg0: Any): Boolean

   Definition Classes

   AnyRef → Any

4.  final def asInstanceOf[T0]: T0

   Definition Classes

   Any

5.  def calculateSpectrogram(waveform: DenseVector[Double], window: DenseVector[Double], frameLength: Int, hopLength: Int, melFilters: DenseMatrix[Double], power: Double, center: Boolean = true, onesided: Boolean = true, melFloor: Double = 1e-10): DenseMatrix[Double]

   Calculates a spectrogram over one waveform using the Short-Time Fourier Transform.

   Calculates a spectrogram over one waveform using the Short-Time Fourier Transform.

   We assume that the waveform has been zero padded beforehand. Currently, this method only supports the mel spectrogram. It uses the breeze implementation of the fourier transform.

   Adapted from huggingface transformer.

   How this works:

   1. The input waveform is split into frames of size frameLength that are partially overlapping by frameLength - hopLength samples.
   2. Each frame is multiplied by the window.
   3. The DFT is taken of each windowed frame.
   4. The resulting rows of the spectrogram are stacked to form the final matrix.

   waveform

   The input waveform to process

   window

   The window to use for each frame

   frameLength

   Length of each frame

   hopLength

   Length to advance in the waveform for each overlapping step

   melFilters

   The mel filters to apply

   power

   Exponent to scale the spectrogram

   center

   Whether to center the waveform and pad reflectively

   onesided

   Whether to only return the positive DFT

   melFloor

   Lowest value to apply for the mel spectrogram

   returns

   Log Mel Spectrogram of the waveform

6.  def clone(): AnyRef

   Attributes

   protected[lang]

   Definition Classes

   AnyRef

   Annotations

   @throws( ... ) @native()

7.  final def eq(arg0: AnyRef): Boolean

   Definition Classes

   AnyRef

8.  def equals(arg0: Any): Boolean

   Definition Classes

   AnyRef → Any

9.  def finalize(): Unit

   Attributes

   protected[lang]

   Definition Classes

   AnyRef

   Annotations

   @throws( classOf[java.lang.Throwable] )

10.  final def getClass(): Class[_]

    Definition Classes

    AnyRef → Any

    Annotations

    @native()

11.  def hashCode(): Int

    Definition Classes

    AnyRef → Any

    Annotations

    @native()

12.  def hertzToMel(freq: Double): Double

    Converts Hertz to Mels.

    Converts Hertz to Mels.

    Uses method proposed by Slaney 1998, adapted from huggingface transformer.

    freq

    Frequency in Hertz

    returns

    Frequency in Mels

13.  final def isInstanceOf[T0]: Boolean

    Definition Classes

    Any

14.  def matrixToFloatArray(matrix: DenseMatrix[Double]): Array[Array[Float]]
15.  def melFilterBank(numFrequencyBins: Int, numMelFilters: Int, minFrequency: Double, maxFrequency: Double, samplingRate: Int): DenseMatrix[Double]

    Creates a matrix to convert a spectrogram to the mel scale.

    Creates a matrix to convert a spectrogram to the mel scale.

    Currently only the filter bank from the Auditory Toolbox for MATLAB (Slaney 1998) is supported.

    Adapted from huggingface transformer.

    numFrequencyBins

    Number of the frequency bins after used for fourier transform

    numMelFilters

    Number of mel filters to generate

    minFrequency

    Lowest frequency in Hz

    maxFrequency

    Highest frequency in Hz

    samplingRate

    Sample rate of the audio

16.  def melToHertz(mels: Double): Double

    Converts Hertz to Mels.

    Converts Hertz to Mels.

    Uses method proposed by Slaney 1998, adapted from huggingface transformer.

    mels

    Frequency in Mels

    returns

    Frequency in Hertz

17.  final def ne(arg0: AnyRef): Boolean

    Definition Classes

    AnyRef

18.  final def notify(): Unit

    Definition Classes

    AnyRef

    Annotations

    @native()

19.  final def notifyAll(): Unit

    Definition Classes

    AnyRef

    Annotations

    @native()

20.  def padReflective(vector: DenseVector[Double], padding: (Int, Int)): DenseVector[Double]

    Pads a vector with the reflection of the vector.

    Pads a vector with the reflection of the vector. It is mirrored on the first and last values of it along each axis.

    This method is adapted from numpy. However, reflections larger than the initial vector are currently not supported.

    vector

    Vector to be padded

    padding

    Padding for the left and right side

    returns

    Vector of size padding._1 + vector.length + padding._2

21.  final def synchronized[T0](arg0: ⇒ T0): T0

    Definition Classes

    AnyRef

22.  def toString(): String

    Definition Classes

    AnyRef → Any

23.  final def wait(): Unit

    Definition Classes

    AnyRef

    Annotations

    @throws( ... )

24.  final def wait(arg0: Long, arg1: Int): Unit

    Definition Classes

    AnyRef

    Annotations

    @throws( ... )

25.  final def wait(arg0: Long): Unit

    Definition Classes

    AnyRef

    Annotations

    @throws( ... ) @native()