We consider source coding of audio signals with the help of a generative model. We use a construction where a waveform is first quantized, yielding a finite bitrate representation. The waveform is then reconstructed by random sampling from a model conditioned on the quantized waveform. The proposed coding scheme is theoretically analyzed. Using SampleRNN as the generative model, we demonstrate that the proposed coding structure provides performance competitive with state-of-the-art source coding tools for specific categories of audio signals.
Dolby Laboratories, Inc.（杜比实验室公司）；Dolby Sweden AB（杜比瑞典公司）
We propose a neural audio generative model, MDCTNet, operating in the perceptually weighted domain of an adaptive modified discrete cosine transform (MDCT). The architecture of the model captures correlations in both time and frequency directions with recurrent layers (RNNs). An audio coding system is obtained by training MDCTNet on a diverse set of fullband monophonic audio signals at 48 kHz sampling, conditioned by a perceptual audio encoder. In a subjective listening test with ten excerpts chosen to be balanced across content types, yet stressful for both codecs, the mean performance of the…
Indiana University（印第安纳大学）；Electronics and Telecommunications Research Institute（韩国电子通信研究院）
Spectral sub-bands do not portray the same perceptual relevance. In audio coding, it is therefore desirable to have independent control over each of the constituent bands so that bitrate assignment and signal reconstruction can be achieved efficiently. In this work, we present a novel neural audio coding network that natively supports a multi-band coding paradigm. Our model extends the idea of compressed skip connections in the U-Net-based codec, allowing for independent control over both core and high band-specific reconstructions and bit allocation. Our system reconstructs the full-band sign…
Indiana University（印第安纳大学）；Electronics and Telecommunications Research Institute（韩国电子通信研究院）
Conventional audio coding technologies commonly leverage human perception of sound, or psychoacoustics, to reduce the bitrate while preserving the perceptual quality of the decoded audio signals. For neural audio codecs, however, the objective nature of the loss function usually leads to suboptimal sound quality as well as high run-time complexity due to the large model size. In this work, we present a psychoacoustic calibration scheme to re-define the loss functions of neural audio coding systems so that it can decode signals more perceptually similar to the reference, yet with a much lower m…
When detecting anomalies in audio, it can often be necessary to consider concept drift: the distribution of the data may drift over time because of dynamically changing environments, and anomalies may become normal as time elapses. We propose to use adaptive Huffman coding for anomaly detection in audio with concept drift. Compared with the existing method of adaptive Gaussian mixture modeling (AGMM), adaptive Huffman coding does not require a priori information about the clusters and can adjust the number of clusters dynamically depending on the amount of variation in the audio. To control th…
This paper introduces a novel neural network-based speech coding system that can process noisy speech effectively. The proposed source-aware neural audio coding (SANAC) system harmonizes a deep autoencoder-based source separation model and a neural coding system so that it can explicitly perform source separation and coding in the latent space. An added benefit of this system is that the codec can allocate a different amount of bits to the underlying sources so that the more important source sounds better in the decoded signal. We target a new use case where the user on the receiver side cares…
Recent speech and audio coding standards such as 3GPP Enhanced Voice Services match the foreseeable needs and requirements in transmission of speech and audio, when using current transmission infrastructure and applications. Trends in Internet-of-Things technology and development in personal digital assistants (PDAs) however begs us to consider future requirements for speech and audio codecs. The opportunities and challenges are here summarized in three concepts: collaboration, unification and privacy. First, an increasing number of devices will in the future be speech-operated, whereby the ab…
University of Erlangen-Nuremberg（埃尔朗根-纽伦堡大学）；International Audio Laboratories（国际音频实验室）
Previous DCASE challenges contributed to an increase in the performance of acoustic scene classification systems. State-of-the-art classifiers demand significant processing capabilities and memory which is challenging for resource-constrained mobile or IoT edge devices. Thus, it is more likely to deploy these models on more powerful hardware and classify audio recordings previously uploaded (or streamed) from low-power edge devices. In such scenario, the edge device may apply perceptual audio coding to reduce the transmission data rate. This paper explores the effect of perceptual audio coding…
There have been several successful deep learning models that perform audio super-resolution. Many of these approaches involve using preprocessed feature extraction which requires a lot of domain-specific signal processing knowledge to implement. Convolutional Neural Networks (CNNs) improved upon this framework by automatically learning filters. An example of a convolutional approach is AudioUNet, which takes inspiration from novel methods of upsampling images. Our paper compares the pre-upsampling AudioUNet to a new generative model that upsamples the signal before using deep learning to trans…
APCodec+: A Spectrum-Coding-Based High-Fidelity and High-Compression-Rate Neural Audio Codec with Staged Training Paradigm
National Engineering Research Center of Speech and Language Information Processing（国家语言语音信息处理工程研究中心）；University of Science and Technology of China（中国科学技术大学）
Audio codecs are typically transform-domain based and efficiently code stationary audio signals, but they struggle with speech and signals containing dense transient events such as applause. Specifically, with these two classes of signals as examples, we demonstrate a technique for restoring audio from coding noise based on generative adversarial networks (GAN). A primary advantage of the proposed GAN-based coded audio enhancer is that the method operates end-to-end directly on decoded audio samples, eliminating the need to design any manually-crafted frontend. Furthermore, the enhancement app…
Zhejiang University（浙江大学）；State Key Laboratory of Media Convergence Production Technology and Systems（媒体融合生产技术与系统国家重点实验室）
Audio super-resolution is the task of constructing a high-resolution (HR) audio from a low-resolution (LR) audio by adding the missing band. Previous methods based on convolutional neural networks and mean squared error training objective have relatively low performance, while adversarial generative models are difficult to train and tune. Recently, normalizing flow has attracted a lot of attention for its high performance, simple training and fast inference. In this paper, we propose WSRGlow, a Glow-based waveform generative model to perform audio super-resolution. Specifically, 1) we integrat…
Sony Computer Science Laboratory（索尼计算机科学实验室）；LTCI, Télécom Paris, Institut Polytechnique de Paris（LTCI，巴黎电信学院，巴黎综合理工学院）
Influenced by the field of Computer Vision, Generative Adversarial Networks (GANs) are often adopted for the audio domain using fixed-size two-dimensional spectrogram representations as the "image data". However, in the (musical) audio domain, it is often desired to generate output of variable duration. This paper presents VQCPC-GAN, an adversarial framework for synthesizing variable-length audio by exploiting Vector-Quantized Contrastive Predictive Coding (VQCPC). A sequence of VQCPC tokens extracted from real audio data serves as conditional input to a GAN architecture, providing step-wise t…
Amsterdam Machine Learning Lab（阿姆斯特丹机器学习实验室）；University of Amsterdam（阿姆斯特丹大学）
Recent advancements in deep generative modeling make it possible to learn prior distributions from complex data that subsequently can be used for Bayesian inference. However, we find that distributions learned by deep generative models for audio signals do not exhibit the right properties that are necessary for tasks like audio source separation using a probabilistic approach. We observe that the learned prior distributions are either discriminative and extremely peaked or smooth and non-discriminative. We quantify this behavior for two types of deep generative models on two audio datasets.
Visually-Guided Sound Source Separation with Audio-Visual Predictive Coding
Xi’an Jiaotong University（西安交通大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Hong Kong Institute of Science & Innovation, Chinese Academy of Sciences（中国科学院香港创新研究院）
Seoul National University（首尔大学）；Kakao Corporation（Kakao公司）
Most modern text-to-speech architectures use a WaveNet vocoder for synthesizing high-fidelity waveform audio, but there have been limitations, such as high inference time, in its practical application due to its ancestral sampling scheme. The recently suggested Parallel WaveNet and ClariNet have achieved real-time audio synthesis capability by incorporating inverse autoregressive flow for parallel sampling. However, these approaches require a two-stage training pipeline with a well-trained teacher network and can only produce natural sound by using probability distillation along with auxiliary…
Sony Computer Science Laboratories（索尼计算机科学实验室）；Télecom Paris（巴黎电信学院）；Institut Polytechnique de Paris（巴黎理工学院）
In this paper, we compare different audio signal representations, including the raw audio waveform and a variety of time-frequency representations, for the task of audio synthesis with Generative Adversarial Networks (GANs). We conduct the experiments on a subset of the NSynth dataset. The architecture follows the benchmark Progressive Growing Wasserstein GAN. We perform experiments both in a fully non-conditional manner as well as conditioning the network on the pitch information. We quantitatively evaluate the generated material utilizing standard metrics for assessing generative models, and…
Recently end-to-end neural audio/speech coding has shown its great potential to outperform traditional signal analysis based audio codecs. This is mostly achieved by following the VQ-VAE paradigm where blind features are learned, vector-quantized and coded. In this paper, instead of blind end-to-end learning, we propose to learn disentangled features for real-time neural speech coding. Specifically, more global-like speaker identity and local content features are learned with disentanglement to represent speech. Such a compact feature decomposition not only achieves better coding efficiency by…
Indiana University（印第安纳大学）；Electronics and Telecommunications Research Institute（韩国电子通信研究院）
An autoencoder-based codec employs quantization to turn its bottleneck layer activation into bitstrings, a process that hinders information flow between the encoder and decoder parts. To circumvent this issue, we employ additional skip connections between the corresponding pair of encoder-decoder layers. The assumption is that, in a mirrored autoencoder topology, a decoder layer reconstructs the intermediate feature representation of its corresponding encoder layer. Hence, any additional information directly propagated from the corresponding encoder layer helps the reconstruction. We implement…
Université de Lorraine（洛林大学）；CNRS（法国国家科学研究中心）；Inria（法国国家信息与自动化研究所）；LORIA（洛林计算机科学及其应用研究所）
Deep latent variable generative models based on variational autoencoder (VAE) have shown promising performance for audiovisual speech enhancement (AVSE). The underlying idea is to learn a VAEbased audiovisual prior distribution for clean speech data, and then combine it with a statistical noise model to recover a speech signal from a noisy audio recording and video (lip images) of the target speaker. Existing generative models developed for AVSE do not take into account the sequential nature of speech data, which prevents them from fully incorporating the power of visual data. In this paper, w…
University of Michigan（密歇根大学）；University of Washington（华盛顿大学）
Neural network-based methods have recently demonstrated state-of-the-art results on image synthesis and super-resolution tasks, in particular by using variants of generative adversarial networks (GANs) with supervised feature losses. Nevertheless, previous feature loss formulations rely on the availability of large auxiliary classifier networks, and labeled datasets that enable such classifiers to be trained. Furthermore, there has been comparatively little work to explore the applicability of GAN-based methods to domains other than images and video. In this work we explore a GAN-based method …
