Data-Free Knowledge Distillation (DFKD) has recently attracted growing attention in the academic community, especially with major breakthroughs in computer vision. Despite promising results, the technique has not been well applied to audio and signal processing. Due to the variable duration of audio signals, it has its own unique way of modeling. In this work, we propose feature-rich audio model inversion (FRAMI), a data-free knowledge distillation framework for general sound classification tasks. It first generates high-quality and feature-rich Mel-spectrograms through a feature-invariant contrastive loss. Then, the hidden states before and after the statistics pooling layer are reused when knowledge distillation is performed on these feature-rich samples. Experimental results on the Urbansound8k, ESC-50, and audioMNIST datasets demonstrate that FRAMI can generate feature-rich samples. Meanwhile, the accuracy of the student model is further improved by reusing the hidden state and significantly outperforms the baseline method.
Audio classification is an active research area with a wide range of applications. Over the past decade, convolutional neural networks (CNNs) have been the de-facto standard building block for end-to-end audio classification models. Recently, neural networks based solely on self-attention mechanisms such as the Audio Spectrogram Transformer (AST) have been shown to outperform CNNs. In this paper, we find an intriguing interaction between the two very different models - CNN and AST models are good teachers for each other. When we use either of them as the teacher and train the other model as th…
This paper introduces the Procedural (audio) Variational autoEncoder (ProVE) framework as a general approach to learning Procedural Audio PA models of environmental sounds with an improvement to the realism of the synthesis while maintaining provision of control over the generated sound through adjustable parameters. The framework comprises two stages: (i) Audio Class Representation, in which a latent representation space is defined by training an audio autoencoder, and (ii) Control Mapping, in which a joint function of static/temporal control variables derived from the audio and a random samp…
Transfer learning is critical for efficient information transfer across multiple related learning problems. A simple, yet effective transfer learning approach utilizes deep neural networks trained on a large-scale task for feature extraction. Such representations are then used to learn related downstream tasks. In this paper, we investigate transfer learning capacity of audio representations obtained from neural networks trained on a large-scale sound event detection dataset. We build and evaluate these representations across a wide range of other audio tasks, via a simple linear classifier tr…
Automated Audio Captioning is a cross-modal task, generating natural language descriptions to summarize the audio clips' sound events. However, grounding the actual sound events in the given audio based on its corresponding caption has not been investigated. This paper contributes an AudioGrounding dataset, which provides the correspondence between sound events and the captions provided in Audiocaps, along with the location (timestamps) of each present sound event. Based on such, we propose the text-to-audio grounding (TAG) task, which interactively considers the relationship between audio pro…
This technical report proposes an audio captioning system for DCASE 2021 Task 6 audio captioning challenge. Our proposed model is based on an encoder-decoder architecture with bi-directional Gated Recurrent Units (BiGRU) using pretrained audio features and sound event detection. A pretrained neural network (PANN) is used to extract audio features and Word2Vec is selected with the aim of extracting word embeddings from the audio captions. To create semantically meaningful captions, we extract sound events from the audio clips and feed the encoder-decoder architecture with sound events in additi…
Text Prompt is Not Enough: Sound Event Enhanced Prompt Adapter for Target Style Audio Generation
SDU-ANU Joint Science College, Shandong University（山东大学SDU-ANU联合科学学院）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Tsinghua University（清华大学）；AI Lab, Tencent（腾讯AI实验室）
Epic-Sounds: A Large-scale Dataset of Actions That Sound
Visual Geometry Group, Department of Engineering Science, University of Oxford, UK（牛津大学视觉几何组）；Department of Computer Science, University of Bristol, UK（布里斯托大学计算机科学系）；CIIRC, Czech Technical University in Prague, Czech Republic（布拉格捷克技术大学CIIRC）
Generative Adversarial Networks (GANs) have achieved excellent audio synthesis quality in the last years. However, making them operable with semantically meaningful controls remains an open challenge. An obvious approach is to control the GAN by conditioning it on metadata contained in audio datasets. Unfortunately, audio datasets often lack the desired annotations, especially in the musical domain. A way to circumvent this lack of annotations is to generate them, for example, with an automatic audio-tagging system. The output probabilities of such systems (so-called "soft labels") carry rich …
Generating Diverse Audio-Visual 360 Soundscapes for Sound Event Localization and Detection
University of Southern California（南加州大学）；Universidad Nacional Autónoma de México（墨西哥国立自治大学）；Universidad Autónoma de Tamaulipas（塔毛利帕斯自治大学）；Queen Mary University of London（伦敦大学玛丽皇后学院）
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Google Research（谷歌研究院）
Recent progress in deep learning has enabled many advances in sound separation and visual scene understanding. However, extracting sound sources which are apparent in natural videos remains an open problem. In this work, we present AudioScope, a novel audio-visual sound separation framework that can be trained without supervision to isolate on-screen sound sources from real in-the-wild videos. Prior audio-visual separation work assumed artificial limitations on the domain of sound classes (e.g., to speech or music), constrained the number of sources, and required strong sound separation or vis…
Improving the performance of on-device audio classification models remains a challenge given the computational limits of the mobile environment. Many studies leverage knowledge distillation to boost predictive performance by transferring the knowledge from large models to on-device models. However, most lack a mechanism to distill the essence of the temporal information, which is crucial to audio classification tasks, or similar architecture is often required. In this paper, we propose a new knowledge distillation method designed to incorporate the temporal knowledge embedded in attention weig…
Teaching Audio Models to Reason: A Unified Framework for Source- and Layer-wise Distillation
Jiutian Artificial Intelligence Research Institute, China Mobile, Beijing, China（九天人工智能研究院，中国移动，北京）；The State Key Laboratory of Multimedia Information Processing, Peking University, Beijing, China（多媒体信息处理国家重点实验室，北京大学，北京）
Audio event detection is a widely studied audio processing task, with applications ranging from self-driving cars to healthcare. In-the-wild datasets such as Audioset have propelled research in this field. However, many efforts typically involve manual annotation and verification, which is expensive to perform at scale. Movies depict various real-life and fictional scenarios which makes them a rich resource for mining a wide-range of audio events. In this work, we present a dataset of audio events called Subtitle-Aligned Movie Sounds (SAM-S). We use publicly-available closed-caption transcript…
Incorporating Ultrasound Tongue Images for Audio-Visual Speech Enhancement through Knowledge Distillation
National Engineering Research Center of Speech and Language Information Processing, University of Science and Technology of China（中国科学技术大学国家语言语音信息处理工程研究中心）
University of Cambridge（剑桥大学）；Jomo Kenyatta University of Agriculture and Technology（乔莫·肯雅塔农业技术大学）
Machine hearing of the environmental sound is one of the important issues in the audio recognition domain. It gives the machine the ability to discriminate between the different input sounds that guides its decision making. In this work we exploit the self-supervised contrastive technique and a shallow 1D CNN to extract the distinctive audio features (audio representations) without using any explicit annotations.We generate representations of a given audio using both its raw audio waveform and spectrogram and evaluate if the proposed learner is agnostic to the type of audio input. We further u…
Knowledge Distillation (KD) is a popular area of research for reducing the size of large models while still maintaining good performance. The outputs of larger teacher models are used to guide the training of smaller student models. Given the repetitive nature of acoustic events, we propose to leverage this information to regulate the KD training for Audio Tagging. This novel KD method, "Intra-Utterance Similarity Preserving KD" (IUSP), shows promising results for the audio tagging task. It is motivated by the previously published KD method: "Similarity Preserving KD" (SP). However, instead of…
Reusing recorded sounds (sampling) is a key component in Electronic Music Production (EMP), which has been present since its early days and is at the core of genres like hip-hop or jungle. Commercial and non-commercial services allow users to obtain collections of sounds (sample packs) to reuse in their compositions. Automatic classification of one-shot instrumental sounds allows automatically categorising the sounds contained in these collections, allowing easier navigation and better characterisation. Automatic instrument classification has mostly targeted the classification of unprocessed i…
Spooky Audio；University of Victoria（维多利亚大学）；McGill University（麦吉尔大学）
We release synth1B1, a multi-modal audio corpus consisting of 1 billion 4-second synthesized sounds, paired with the synthesis parameters used to generate them. The dataset is 100x larger than any audio dataset in the literature. We also introduce torchsynth, an open source modular synthesizer that generates the synth1B1 samples on-the-fly at 16200x faster than real-time (714MHz) on a single GPU. Finally, we release two new audio datasets: FM synth timbre and subtractive synth pitch. Using these datasets, we demonstrate new rank-based evaluation criteria for existing audio representations. Fin…
University of Surrey（萨里大学）；Tencent AI Lab（腾讯人工智能实验室）；Fictional University（虚构大学）；University of the Imagination（想象大学）
The Detection and Classification of Acoustic Scenes and Events (DCASE) 2019 challenge focuses on audio tagging, sound event detection and spatial localisation. DCASE 2019 consists of five tasks: 1) acoustic scene classification, 2) audio tagging with noisy labels and minimal supervision, 3) sound event localisation and detection, 4) sound event detection in domestic environments, and 5) urban sound tagging. In this paper, we propose generic cross-task baseline systems based on convolutional neural networks (CNNs). The motivation is to investigate the performance of a variety of models across s…
Google Research（谷歌研究院）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
We introduce AudioScopeV2, a state-of-the-art universal audio-visual on-screen sound separation system which is capable of learning to separate sounds and associate them with on-screen objects by looking at in-the-wild videos. We identify several limitations of previous work on audio-visual on-screen sound separation, including the coarse resolution of spatio-temporal attention, poor convergence of the audio separation model, limited variety in training and evaluation data, and failure to account for the trade off between preservation of on-screen sounds and suppression of off-screen sounds. W…
Many state-of-the-art systems for audio tagging and sound event detection employ convolutional recurrent neural architectures. Typically, they are trained in a mean teacher setting to deal with the heterogeneous annotation of the available data. In this work, we present a thorough analysis of how changing the temporal resolution of these convolutional recurrent neural networks - which can be done by simply adapting their pooling operations - impacts their performance. By using a variety of evaluation metrics, we investigate the effects of adapting this design parameter under several sound reco…
Although several research works have been reported on audio-visual sound source localization in unconstrained videos, no datasets and metrics have been proposed in the literature to quantitatively evaluate its performance. Defining the ground truth for sound source localization is difficult, because the location where the sound is produced is not limited to the range of the source object, but the vibrations propagate and spread through the surrounding objects. Therefore we propose a new concept, Sounding Object, to reduce the ambiguity of the visual location of sound, making it possible to ann…
National Institute of Advanced Industrial Science and Technology（日本产业技术综合研究所）；Waseda University（早稻田大学）
Detecting sound source objects within visual observation is important for autonomous robots to comprehend surrounding environments. Since sounding objects have a large variety with different appearances in our living environments, labeling all sounding objects is impossible in practice. This calls for self-supervised learning which does not require manual labeling. Most of conventional self-supervised learning uses monaural audio signals and images and cannot distinguish sound source objects having similar appearances due to poor spatial information in audio signals. To solve this problem, thi…
We study the merit of transfer learning for two sound recognition problems, i.e., audio tagging and sound event detection. Employing feature fusion, we adapt a baseline system utilizing only spectral acoustic inputs to also make use of pretrained auditory and visual features, extracted from networks built for different tasks and trained with external data. We perform experiments with these modified models on an audiovisual multi-label data set, of which the training partition contains a large number of unlabeled samples and a smaller amount of clips with weak annotations, indicating the clip-l…
We propose a method of separating a desired sound source from a single-channel mixture, based on either a textual description or a short audio sample of the target source. This is achieved by combining two distinct models. The first model, SoundWords, is trained to jointly embed both an audio clip and its textual description to the same embedding in a shared representation. The second model, SoundFilter, takes a mixed source audio clip as an input and separates it based on a conditioning vector from the shared text-audio representation defined by SoundWords, making the model agnostic to the co…
Audio denoising has been explored for decades using both traditional and deep learning-based methods. However, these methods are still limited to either manually added artificial noise or lower denoised audio quality. To overcome these challenges, we collect a large-scale natural noise bird sound dataset. We are the first to transfer the audio denoising problem into an image segmentation problem and propose a deep visual audio denoising (DVAD) model. With a total of 14,120 audio images, we develop an audio ImageMask tool and propose to use a few-shot generalization strategy to label these imag…
