Huawei Technologies Co., Limited

Huawei Technologies Duesseldorf Gmbh

University of Cambridge

Milos Markovic

Panji Setiawan

Karim Helwani

Florian Eyben

Simon J Godsill

Herbert Buchner

Jürgen Geiger

Björn Schuller

Andrea Crespi

An apparatus and a method for acoustic scene classification of a block of audio samples are provided. The block is partitioned into frames in the time domain. For each respective frame of a plurality of frames of the block, a change measure between the respective frame and a preceding frame of the block is calculated. The respective frame is assigned, based on the calculated change measure, to one of a set of short-event frames, a set of long-event frames, and a set of background frames. The feature vector is determined based on a feature computed from one or more of the set of short-event frames, the set of long-event frames, and the set of background frames.

Segmentation-based feature extraction for acoustic scene classification

An input audio signal comprises a plurality of input audio channels. A KLT-based pre-processor transforms the plurality of input audio channels into a plurality of eigenchannels and provides metadata associated with the plurality of eigenchannels. Each eigenchannel is associated with an eigenvalue and an eigenvector. The metadata allows reconstructing the plurality of input audio channels on the basis of the plurality of eigenchannels. A selector selects a subset of the plurality of eigenvectors corresponding to a plurality of selected eigenchannels on the basis of a geometric mean of the eigenvalues. An eigenchannel encoder encodes the plurality of selected eigenchannels. A metadata encoder encodes the metadata.

Apparatuses and methods for encoding and decoding a multichannel audio signal

A signal processing apparatus and method are provided for separating a plurality of mixture signals from a MIMO system to iteratively obtain a plurality of output signals. The plurality of mixture signals are a response of the MIMO system to a plurality of source signals. The signal processing apparatus comprises a plurality of blind source separators including a first blind source separator based on a first blind source separation technique or algorithm and a second blind source separator based on a second blind source separation technique or algorithm, wherein the first blind source separator is configured to compute a first plurality of preliminary output signals on the basis of a first set of coefficients describing the MIMO system and wherein the second blind source separator is configured to compute a second plurality of preliminary output signals on the basis of a second set of coefficients describing the MIMO system.

Signal processing apparatus and method

A device having an input that is configured to receive a sound signal captured from an acoustic environment where the device is located, is provided. The device also includes a signal processor that is configured to obtain a reverberation signature from the sound signal and classify the acoustic environment as being of a particular type in dependence on that reverberation signature. It also includes a controller configured to control an operation of the device in dependence on the classification of the acoustic environment. By using the reverberation signature to classify the environment, the device it is not dependent on a particular type of signal having to be present in the environment at the time when the sound signal is captured. The reverberation properties of the environment will be embodied by any sound signal captured from that environment.

Device and method for classifying an acoustic environment

Growing community of inventors

Munich, Germany

Milos Markovic