Cisco Technology, Inc.

Intel Corporation

Lsi Corporation

Agere Systems Inc.

Cadence Design Systems, Inc.

Babblelabs LLC

Avago Technologies General IP (Singapore) Pte. Ltd.

Samer Lutfi Hijazi

Kameran Azadet

Christopher Rowen

Xuehong Mao

Raul Alejandro Casas

Albert Molina

Joseph Williams

Jian-Guo Chen

Ramon Sanchez

Joseph H Othmer

Dror E Maydan

Piyush Kaul

Rishi Kumar

Sunitha Kopparthi

Marcin Ciolek

Yusuf Ziya Isik

Kamil Krzysztof Wojcicki

Raúl Alejandro Casas

Ivana M Balic

Amir Salah Abdelsamie Abdelwahed

Keith Griffin

Hua Mu

Mihailo Kolundzija

Michal Sulewski

Dror Maydan

Meng-Lin Yu

Carl Murray

David Guoqing Zhang

Erik S Panu

Ramon Sanchez Perez

Himanshu Charaya

Joe H Othmer

Erik Panu

In one embodiment, an illustrative method herein may comprise: receiving, at a receiver, an audio codec stream; determining, by the receiver, a length of time associated with a look-ahead buffer of the audio codec stream; inputting, by the receiver, the audio codec stream into an audio enhancement model trained with one or more audio enhancements to cause the audio enhancement model to apply the one or more audio enhancements to the audio codec stream to generate an enhanced audio codec stream within the length of time associated with the look-ahead buffer of the audio codec stream; and outputting, by the receiver, the enhanced audio codec stream.

Receiver side noise reduction and speech enhancement for PSTN callers

In one example embodiment, speech signals are received from a user during a communication session. The received speech signals contain noise including speech of other individuals. The received speech signals are transformed by a machine learning model to produce transformed speech signals corresponding to the received speech signals with a reduced amount of the noise. The machine learning model is trained with speech of the user satisfying a noise threshold and collected during one or more communication sessions.

Transforming speech signals to attenuate speech of competing individuals and other noise

In some aspects, the techniques described herein relate to a method including: encoding a current data portion to generate an encoded current data portion for inclusion in a data packet; encoding, based upon content of the current data portion, a forward error correction data portion for a previous data portion to generate an encoded forward error correction data portion; generating the data packet including the encoded current data portion and the encoded forward error correction data portion; and providing the data packet to a receiver.

Bandwidth utilization techniques for in-band redundant data

A method comprises: vector quantizing input vectors representative of audio into an original sequence including indices of codewords of a codebook; generating candidate sequences including the indices of the codewords of the codebook by evaluating, for each candidate sequence, transition costs for transitions between the indices based on (i) transition probabilities of the transitions, and (ii) distances between the codewords represented by the indices and the input vectors that corresponds to the indices; determining a preferred candidate sequence of the candidate sequences to replace the original sequence based on the transition costs for each candidate sequence; and transmitting the preferred candidate sequence in place of the original sequence.

Vector quantizer correction for audio codec system

A method comprises: storing n dimension (nD) dictionaries (nD dictionaries) where n decreases from a highest dimension to a lowest dimension, each nD dictionary including codewords for sequences of n symbols that are of a limited number that is less than all possible sequences of n symbols; storing key blocks for corresponding ones of the nD dictionaries, each key block configured with keys that map sequences of n−1 symbols to dictionaries of a corresponding one of the nD dictionaries that includes the codewords; receiving a sequence of symbols that represent indices of codevectors of a vector quantizer codebook that are representative of audio; determining a codeword using the key blocks and the nD dictionaries; and encoding a current symbol of the sequence of symbols using the codeword.

Reduced multidimensional indices compression for audio codec system

Presented herein are techniques in which a first device connects to a communication session in which a plurality of devices communicates. The plurality of devices includes the first device and a second device. The first device outputs first audio that includes a first audio watermark associated with the communication session and the second device outputs second audio that includes a second audio watermark associated with the communication session. The first device detects the second audio watermark in the second audio outputted by the second device and one or more actions are performed in response to detecting the second audio watermark.

Audio watermarking to prevent meeting howl

In some aspects, the techniques described herein relate to a method including: obtaining data to be compressed; determining a distance between the data to be compressed and each codeword of a plurality of codewords; selecting a predetermined number of codewords of the plurality of codewords based on the distance between the data to be compressed and each of the predetermined number of codewords; and generating compressed data, where the compressed data includes an indication of the predetermined number of codewords of the plurality of codewords.

Computationally efficient and bitrate scalable soft vector quantization

Systems and methods are disclosed for audio enhancement. For example, methods may include accessing audio data; determining a window of audio samples based on the audio data; inputting the window of audio samples to a classifier to obtain a classification, in which the classifier includes a neural network and the classification takes a value from a set of multiple classes of audio; selecting, based on the classification, an audio enhancement network from a set of multiple audio enhancement networks; applying the selected audio enhancement network to the window of audio samples to obtain an enhanced audio segment, in which the selected audio enhancement network includes a neural network that has been trained using audio signals of a type associated with the classification; and storing, playing, or transmitting an enhanced audio signal based on the enhanced audio segment.

Data driven audio enhancement

Systems and methods are disclosed for audio group identification for conferencing. For example, methods may include joining a conference call using a network interface; accessing an audio signal that has been captured using a microphone; detecting a control signal in the audio signal; and, responsive to detection of the control signal, invoking modification of an audio path of the conference call.

Audio group identification for conferencing

Systems and methods are disclosed for data driven radio enhancement. For example, methods may include demodulating a radio signal to obtain a demodulated audio signal; determining a window of audio samples based on the demodulated audio signal; applying an audio enhancement network to the window of audio samples to obtain an enhanced audio segment, in which the audio enhancement network includes a machine learning network that has been trained using demodulated audio signals derived from radio signals; and storing, playing, or transmitting an enhanced audio signal based on the enhanced audio segment.

Data driven radio enhancement

Some embodiments perform, in a multi-layer neural network in a computing device, a convolution operation on input feature maps with multiple convolutional filters. The convolutional filters have multiple filter precisions. In other embodiments, electronic design automation (EDA) systems, methods, and computer-readable media are presented for adding such a multi-layer neural network into an integrated circuit (IC) design.

Filtering in trainable networks

Some embodiments perform, in a multi-layer neural network in a computing device, optimization of the multi-layer neural network, for example by making a convolutional change with a first plurality of convolutional filters, or by making a connection change of a first plurality of convolutional filters. In other embodiments, electronic design automation (EDA) systems, methods, and computer-readable media are presented for adding such a multi-layer neural network into an integrated circuit (IC) design.

Complexity optimization of trainable networks

Growing community of inventors

San Jose, CA, United States of America

Samer Lutfi Hijazi