Electronics and Telecommunications Research Institute

Indiana University

Yonsei University

Gwangju Institute of Science and Technology

Kyungpook National University Industry-academic Cooperation Foundation

Woo-taek Lim

Tae Jin Lee

Seung Kwon Beack

Jongmo Sung

Inseon Jang

Mi Suk Lee

Hui Yong Kim

Young Ho Jeong

Minje Kim

Sang Won Suh

Byeongho Cho

Youngju Cheon

Soojoong Hwang

Sangwook Han

Jin Soo Choi

Jihyun Lee

Chanwoo Lee

Jae-hyoun Yoo

Jongwon Shin

Hong-Goo Kang

Soo Young Park

Young Cheol Park

Jong Won Shin

Seokjin Lee

Minhan Kim

Jong-Won Seok

Haici Yang

Seungmin Shin

Joon Byun

Yunsu Kim

Hyungseob Lim

A method of encoding a speech signal includes predicting a feature vector of each of a plurality of frames included in the speech signal based on a ground-truth feature vector of a previous frame of each of the plurality of frames, calculating a residual signal corresponding to each of the plurality of frames based on a ground-truth feature vector of each of the plurality of frames and a predicted feature vector of each of the plurality of frames, and generating a bitstring corresponding to each of the plurality of frames by quantizing the residual signal.

Speech coding method and apparatus for performing the same

Disclosed is an apparatus and method for audio encoding/decoding that is robust against coding distortion in a transition section. An audio encoding method includes outputting a frequency domain signal by time-to-frequency (T/F) transform of an input signal, outputting a frequency domain residual signal in which a frequency axis envelope is removed from the frequency domain signal by applying frequency domain noise shaping (FDNS) encoding to the frequency domain signal, outputting a time domain residual signal in which a time axis envelope is removed by performing linear prediction coefficient (LPC) analysis based on the frequency domain residual signal, and quantizing and transmitting the time domain residual signal.

Apparatus and method for audio encoding/decoding robust to transition segment encoding distortion

An encoding method, a decoding method, an encoder for performing the encoding method, and a decoder for performing the decoding method are provided. The encoding method includes outputting LP coefficients bitstream and a residual signal by performing an LP analysis on an input signal, outputting a first latent signal obtained by encoding a periodic component of the residual signal, a second latent signal obtained by encoding a non-periodic component of the residual signal, and a weight vector for each of the first latent signal and the second latent signal, using a first neural network module, and outputting a first bitstream obtained by quantizing the first latent signal, a second bitstream obtained by quantizing the second latent signal, and a weight bitstream obtained by quantizing the weight vector, using a quantization module.

Encoding method, decoding method, encoder for performing encoding method, and decoder for performing decoding method

Provided is a method and apparatus for designing and testing an audio codec using quantization based on white noise modeling. A neural network-based audio encoder design method includes generating a quantized latent vector and a reconstructed signal corresponding to an input signal by using a white noise modeling-based quantization process, computing a total loss for training a neural network-based audio codec, based on the input signal, the reconstruction signal, and the quantized latent vector, training the neural network-based audio codec by using the total loss, and validating the trained neural network-based audio codec to select the best neural network-based audio codec.

Method and apparatus for designing and testing audio codec by using white noise modeling

An audio signal encoding and decoding method using a neural network model, and an encoder and decoder for performing the same are disclosed. A method of encoding an audio signal using a neural network model, the method may include identifying an input signal, generating a quantized latent vector by inputting the input signal into a neural network model encoding the input signal, and generating a bitstream corresponding to the quantized latent vector, wherein the neural network model may include i) a feature extraction layer generating a latent vector by extracting a feature of the input signal, ii) a plurality of downsampling blocks downsampling the latent vector, and iii) a plurality of quantization blocks performing quantization of a downsampled latent vector.

Audio signal encoding and decoding method using a neural network model to generate a quantized latent vector, and encoder and decoder for performing the same

Provided is an encoding method according to various example embodiments and an encoder performing the method. The encoding method includes outputting a linear prediction (LP) coefficients bitstream and a residual signal by performing a linear prediction analysis on an input signal, outputting a first latent signal obtained by encoding a periodic component of the residual signal, using a first neural network module, outputting a first bitstream obtained by quantizing the first latent signal, using a quantization module, outputting a second latent signal obtained by encoding an aperiodic component of the residual signal, using the first neural network module, and outputting a second bitstream obtained by quantizing the second latent signal, using the quantization module, wherein the aperiodic component of the residual signal is calculated based on a periodic component of the residual signal decoded from the quantized first latent signal output by de-quantizing the first bitstream.

Methods of encoding and decoding, encoder and decoder performing the methods

A method of generating a residual signal performed by an encoder includes identifying an input signal including an audio sample, generating a first residual signal from the input signal using linear predictive coding (LPC), generating a second residual signal having a less information amount than the first residual signal by transforming the first residual signal, transforming the second residual signal into a frequency domain, and generating a third residual signal having a less information amount than the second residual signal from the transformed second residual signal using frequency-domain prediction (FDP) coding.

Method of generating residual signal, and encoder and decoder performing the method

An audio signal encoding and decoding method using a neural network model, a method of training the neural network model, and an encoder and decoder performing the methods are disclosed. The encoding method includes computing the first feature information of an input signal using a recurrent encoding model, computing an output signal from the first feature information using a recurrent decoding model, calculating a residual signal by subtracting the output signal from the input signal, computing the second feature information of the residual signal using a nonrecurrent encoding model, and converting the first feature information and the second feature information to a bitstream.

Methods of encoding and decoding audio signal using neural network model, and devices for performing the methods

Disclosed are a method of encoding and decoding an audio signal and an encoder and a decoder performing the method. The method of encoding an audio signal includes identifying an input signal, and generating a bitstring of each encoding layer by applying, to the input signal, an encoding model including a plurality of successive encoding layers that encodes the input signal, in which a current encoding layer among the encoding layers is trained to generate a bitstring of the current encoding layer by encoding an encoded signal which is a signal encoded in a previous encoding layer and quantizing an encoded signal which is a signal encoded in the current encoding layer.

Audio signal encoding and decoding method, and encoder and decoder performing the methods

An audio encoding/decoding apparatus and method using vector quantized residual error features are disclosed. An audio signal encoding method includes outputting a bitstream of a main codec by encoding an original signal, decoding the bitstream of the main codec, determining a residual error feature vector from a feature vector of a decoded signal and a feature vector of the original signal, and outputting a bitstream of additional information by encoding the residual error feature vector.

Audio encoding/decoding apparatus and method using vector quantized residual error feature

Disclosed are methods of encoding and decoding an audio signal using side information, and an encoder and a decoder for performing the methods. The method of encoding an audio signal using side information includes identifying an input signal, the input signal being an original audio signal, extracting side information from the input signal using a learning model trained to extract side information from a feature vector of the input signal, encoding the input signal, and generating a bitstream by combining the encoded input signal and the side information.

Methods of encoding and decoding audio signal using side information, and encoder and decoder for performing the methods

An audio signal encoding and decoding method using a learning model, a training method of the learning model, and an encoder and decoder that perform the method, are disclosed. The audio signal decoding method may include extracting a first residual signal and a first linear prediction coefficient by decoding a bitstream received from an encoder, generating a first audio signal from the first residual signal using the first linear prediction coefficient, generating a second linear prediction coefficients and a second residual signal from the first audio signal, obtaining a third linear prediction coefficient by inputting the second linear prediction coefficient into a trained learning model, and generating a second audio signal from the second residual signal using the third linear prediction coefficient.

Audio signal encoding and decoding method using learning model, training method of learning model, and encoder and decoder that perform the methods

Methods of encoding and decoding a speech signal using a neural network model that recognizes sound sources, and encoding and decoding apparatuses for performing the methods are provided. A method of encoding a speech signal includes identifying an input signal for a plurality of sound sources; generating a latent signal by encoding the input signal; obtaining a plurality of sound source signals by separating the latent signal for each of the plurality of sound sources; determining a number of bits used for quantization of each of the plurality of sound source signals according to a type of each of the plurality of sound sources; quantizing each of the plurality of sound source signals based on the determined number of bits; and generating a bitstream by combining the plurality of quantized sound source signals.

Methods of encoding and decoding speech signal using neural network model recognizing sound sources, and encoding and decoding apparatuses for performing the same

Disclosed is an apparatus and method for augmenting training data using a notch filter. The method may include obtaining original data, and obtaining training data having a modified frequency component from the original data by filtering the original data using a filter configured to remove a component of a predetermined frequency band.

Apparatus and method for augmenting training data using notch filter

Disclosed are methods of encoding and decoding an audio signal, and an encoder and a decoder for performing the methods. The method of encoding an audio signal includes identifying an input signal corresponding to a low frequency band of the audio signal, windowing the input signal, generating a first latent vector by inputting the windowed input signal to a first encoding model, transforming the windowed input signal into a frequency domain, generating a second latent vector by inputting the transformed input signal to a second encoding model, generating a final latent vector by combining the first latent vector and the second latent vector, and generating a bitstream corresponding to the final latent vector.

Growing community of inventors

Daejeon, South Korea

Woo-taek Lim