Changes the aggressiveness of the noise suppression method. |mode| = 0 is mild (6dB), |mode| = 1 is medium (10dB) and |mode| = 2 is aggressive (15dB).

For a normal distribution, the probability of |input| is calculated and returned (in Q20). The formula for normal distributed probability is 1 / s *

Set aggressiveness mode

estimate noise spectrum by using the front self.noise_frames frames :return: noise spectrum

This is a GSM function, for concrete processors she may be replaced for same processor's optimized function (norm_l)

Berouti spectral subtraction :return: enhanced speech

Compares the |binary_vector| with all rows of the |binary_matrix| and counts per row the number of times they have the same value. Inputs: - binary_v

Estimate prior SNR decision-directed and compute DD based Wiener Filter. Input: |magn| is the signal magnitude spectrum estimate. Output: |theFilter|

Compute prior and post SNR based on quantile noise estimation. Compute DD estimate of prior SNR. Inputs: |magn| is the signal magnitude spectrum estim

Compute the difference measure between input spectrum and a template/learned noise spectrum. |magnIn| is the input spectrum. The reference/template sp

Compute spectral flatness on input spectrum. |magnIn| is the magnitude spectrum. Spectral flatness is returned in self->featureData[0].

Calculates the energy of a buffer. Inputs: |buffer| is the buffer over which the energy is calculated. |length| is the length of the buffer. Returns t

Update the noise features. Inputs: |magn| is the signal magnitude spectrum estimate. |updateParsFlag| is an update flag for parameters.

Get address of region(s) from which we can read data. If the region is contiguous, |data_ptr_bytes_2| will be zero. If non-contiguous, |data_ptr_bytes

Calculates the probabilities for both speech and background noise using Gaussian Mixture Models (GMM). A hypothesis-test is performed to decide which

High pass filtering, with a cut-off frequency at 80 Hz, if the |data_in| is sampled at 500 Hz. - data_in [i] : Input audio data sampled at 500

Validates the |candidate_delay|, estimated in WebRtc_ProcessBinarySpectrum(), based on a mix of counting concurring hits with a modified histogram of

Transforms the signal from frequency to time domain. Inputs: |real| is the real part of the frequency domain. |imag| is the imaginary part of the freq

Calculates the mean recursively. Same version as WebRtc_MeanEstimatorFix(), but for float. Inputs: - new_value : New additional value. -

Estimate noise.

An s16 x s32 -> s32 multiplication that's allowed to overflow. (It's still undefined behavior, so not a good idea; this just makes UBSan ignore the vi