오디오 처리

Design an autoencoder neural network to perform anomaly detection for machine sounds using unsupervised learning.

Perform speech enhancement using a pretrained filter and sum network (FaSNet) with ambisonic data.

Perform 3-D sound event localization and detection using a pretrained deep learning model.

Develop an x-vector system to perform speaker recognition.

Speaker diarization is the process of partitioning an audio signal into segments according to speaker identity. It answers the question "who spoke when" without prior knowledge of the speakers and, depending on the application, without prior knowledge of the number of speakers.

이 예제에서는 오디오에서 음성 명령의 존재 여부를 감지하는 딥러닝 모델을 훈련시키는 방법을 보여줍니다. 이 예제에서는 Speech Commands Dataset[1]을 사용하여 컨벌루션 신경망이 명령 세트를 인식하도록 훈련시킵니다.

Identify a keyword in noisy speech using a deep learning network. In particular, the example uses a Bidirectional Long Short-Term Memory (BiLSTM) network and mel frequency cepstral coefficients (MFCC).

이 예제에서는 딥러닝 신경망을 사용하여 음성 신호의 잡음을 제거하는 방법을 다룹니다. 이 예제에서는 동일한 작업에 적용된 두 가지 유형의 신경망, 즉 완전 연결 신경망과 컨벌루션 신경망을 비교합니다.

Train and use a generative adversarial network (GAN) to generate sounds.

In this example, you perform batch and streaming voice activity detection (VAD) in a low SNR environment using a pretrained deep learning model. For details about the model and how it was trained, see Train Voice Activity Detection in Noise Model Using Deep Learning (Audio Toolbox).

Illustrates a simple speech emotion recognition (SER) system using a BiLSTM network. You begin by downloading the data set and then testing the trained network on individual files. The network was trained on a small German-language database [1].

Create a multi-model late fusion system for acoustic scene recognition. The example trains a convolutional neural network (CNN) using mel spectrograms and an ensemble classifier using wavelet scattering. The example uses the TUT dataset for training and evaluation [1].

Use an end-to-end deep learning network for speaker-independent speech separation.

Develop a deep learning model to detect faults in an air compressor and package the system to operate on streaming data.

Perform audio event classification on Raspberry Pi^® using the YAMNet pretrained deep neural network from the TensorFlow™ Lite library.

Demonstrates code generation for keyword spotting using a Bidirectional Long Short-Term Memory (BiLSTM) network and mel frequency cepstral coefficient (MFCC) feature extraction on Raspberry Pi™. MATLAB® Coder™ with Deep Learning Support enables the generation of a standalone executable (.elf) file on Raspberry Pi. Communication between MATLAB® (.mlx) file and the generated executable file occurs over asynchronous User Datagram Protocol (UDP). The incoming speech signal is displayed using a timescope. A mask is shown as a blue rectangle surrounding spotted instances of the keyword, YES. For more details on MFCC feature extraction and deep learning network training, visit Keyword Spotting in Noise Using MFCC and LSTM Networks (Audio Toolbox).

Deploy feature extraction and a convolutional neural network (CNN) for speech command recognition on Intel® processors. To generate the feature extraction and network code, you use MATLAB® Coder™ and the Intel® Math Kernel Library for Deep Neural Networks (MKL-DNN). In this example, the generated code is a MATLAB executable (MEX) function, which is called by a MATLAB script that displays the predicted speech command along with the time domain signal and auditory spectrogram. For details about audio preprocessing and network training, see Train Speech Command Recognition Model Using Deep Learning (Audio Toolbox).

Generate a MATLAB^® standalone executable for acoustics-based machine fault recognition.

Deploy feature extraction and a convolutional neural network for speech command recognition on Raspberry Pi.