【1】 Multiple Hankel matrix rank minimization for audio inpainting标题：用于音频修复的多Hankel矩阵最小秩算法链接：http://arxiv.org/pdf/2303.18023v1
作者：Pavel Záviška,Pavel Rajmic,Ondřej Mokrý机构：Brno University of Technology, Dept. of Telecommunications, Technická , Brno, Czech Republic摘要：Sasaki等人（2018）提出了一种有效的音频去唇算法，该算法基于从时域信号块构建的Hankel结构矩阵的属性。我们采用他们的方法来解决音频修复问题，其中样本在信号中丢失。我们分析了算法，并提供修改，其中一些导致性能的提高。总的来说，事实证明，新的算法表现相当不错的语音信号，但他们没有竞争力的情况下，音乐信号。摘要：Sasaki et al. (2018) presented an efficient audio declipping algorithm, based on the properties of Hankel-structured matrices constructed from time-domain signal blocks. We adapt their approach to solve the audio inpainting problem, where samples are missing in the signal. We analyze the algorithm and provide modifications, some of them leading to an improved performance. Overall, it turns out that the new algorithms perform reasonably well for speech signals but they are not competitive in the case of music signals.【2】 Evaluation of Noise Reduction Methods for Sentence Recognition by Sinhala Speaking Listeners标题：僧伽罗语听者句子识别降噪方法的评价链接：http://arxiv.org/pdf/2303.17829v1
作者：M. B. M. Gunawardhana,U. K. W. D. M. R. N. C. Navanjana,H. D. G. Fernando,R. M. N. Upeksha,A. C. De Silva摘要：降噪是研究人员多年来试图解决的助听器的一个关键方面。大多数降噪算法都是使用英语语音材料进行评估的。英语和僧伽罗语的语言特征有许多不同之处，如音节结构不同、元音时长不同等。小波变换和自适应滤波在助听器降噪中有着广泛的应用。本文比较了10个小波族的小波变换的性能与软阈值和硬阈值方法对自适应滤波器的归一化最小均方（NLMS），最小均方（LMS），平均归一化最小均方（ANLMS），递归最小二乘（RLS），自适应滤波平均（AFA）优化算法以及倒谱和基于能量的语音活动检测（VAD）算法。使用客观指标进行绩效评估;信噪比（SNR）和语音质量的感知评估（PESQ）以及主观度量;平均意见评分（MOS）。NOIZEUS数据库由得克萨斯大学，达拉斯和新成立的僧伽罗语音频数据库被用于评估。摘要：Noise reduction is a critical aspect of hearing aids that researchers trying to solve over the years. Most of the noise reduction algorithms are evaluated using English Speech Material. There are many differences between the linguistic features of English and Sinhala languages, such as different syllable structures and different vowel duration. Both wavelet transformation and adaptive filtering have been widely used for noise reduction in hearing aids. This paper compares the performance of wavelet transformation of ten wavelet families with soft and hard thresholding methods against adaptive filters with Normalized Least Mean Square (NLMS), Least Mean Square (LMS), Average Normalized Least Mean Square (ANLMS), Recursive Least Square (RLS), and Adaptive Filtering Averaging (AFA) optimization algorithms along with cepstral and energy-based voice activity detection (VAD) algorithms. The performance evaluation is done using objective metrics; Signal to Noise Ratio (SNR) and Perceptual Evaluation of Speech Quality (PESQ) and a subjective metric; Mean Opinion Score (MOS). The NOIZEUS database by the University of Texas, Dallas and a newly formed Sinhala language audio database were used for the evaluation.【3】 The Edinburgh International Accents of English Corpus: Towards the Democratization of English ASR标题：英语语料库的爱丁堡国际口音：走向英语ASR的民主化链接：http://arxiv.org/pdf/2303.18110v1
作者：Ramon Sanabria,Nikolay Bogoychev,Nina Markl,Andrea Carmantini,Ondrej Klejch,Peter Bell机构：School of Informatics, The University of Edinburgh备注：Accepted to IEEE ICASSP 2023摘要：None摘要：English is the most widely spoken language in the world, used daily by millions of people as a first or second language in many different contexts. As a result, there are many varieties of English. Although the great many advances in English automatic speech recognition (ASR) over the past decades, results are usually reported based on test datasets which fail to represent the diversity of English as spoken today around the globe. We present the first release of The Edinburgh International Accents of English Corpus (EdAcc). This dataset attempts to better represent the wide diversity of English, encompassing almost 40 hours of dyadic video call conversations between friends. Unlike other datasets, EdAcc includes a wide range of first and second-language varieties of English and a linguistic background profile of each speaker. Results on latest public, and commercial models show that EdAcc highlights shortcomings of current English ASR models. The best performing model, trained on 680 thousand hours of transcribed data, obtains an average of 19.7% word error rate (WER) -- in contrast to the 2.7% WER obtained when evaluated on US English clean read speech. Across all models, we observe a drop in performance on Indian, Jamaican, and Nigerian English speakers. Recordings, linguistic backgrounds, data statement, and evaluation scripts are released on our website (https: groups.inf.ed.ac.uk edacc ) under CC-BY-SA license.【4】 Unsupervised Anomaly Detection and Localization of Machine Audio: A GAN-based Approach标题：机器音频的非监督异常检测与定位：一种基于GAN的方法链接：http://arxiv.org/pdf/2303.17949v1
作者：Anbai Jiang,Wei-Qiang Zhang,Yufeng Deng,Pingyi Fan,Jia Liu机构：BNRist, Department of Electronic Engineering, Tsinghua University, Beijing, China, Tsinghua AI Plus, Beijing, China备注：Accepted by ICASSP 2023摘要：机器异常的自动检测仍然是机器学习的挑战。我们相信生成对抗网络（GAN）的能力适合机器音频异常检测的需要，但以前的工作很少对此进行研究。在本文中，我们提出了AEGAN-AD，一个完全无监督的方法，其中生成器（也是一个自动编码器）进行训练，以重建输入频谱图。它指出，重建的去噪性质贬低其容量。因此，鉴别器被重新设计以在训练阶段和检测阶段期间辅助生成器。AEGAN-AD在DCASE 2022挑战任务2数据集上的性能展示了五种机器类型的最新结果。一种新的异常定位方法也进行了研究。源代码可从以下网址获得：www.github.com jianganbai AEGAN-AD摘要：Automatic detection of machine anomaly remains challenging for machine learning. We believe the capability of generative adversarial network (GAN) suits the need of machine audio anomaly detection, yet rarely has this been investigated by previous work. In this paper, we propose AEGAN-AD, a totally unsupervised approach in which the generator (also an autoencoder) is trained to reconstruct input spectrograms. It is pointed out that the denoising nature of reconstruction deprecates its capacity. Thus, the discriminator is redesigned to aid the generator during both training stage and detection stage. The performance of AEGAN-AD on the dataset of DCASE 2022 Challenge TASK 2 demonstrates the state-of-the-art result on five machine types. A novel anomaly localization method is also investigated. Source code available at: www.github.com jianganbai AEGAN-AD【5】 Dialog act guided contextual adapter for personalized speech recognition标题：用于个性化语音识别的对话动作制导的上下文适配器链接：http://arxiv.org/pdf/2303.17799v1
作者：Feng-Ju Chang,Thejaswi Muniyappa,Kanthashree Mysore Sathyendra,Kai Wei,Grant P. Strimel,Ross McGowan机构：Alexa, Amazon, USA备注：Accepted at ICASSP 2023摘要：None摘要：Personalization in multi-turn dialogs has been a long standing challenge for end-to-end automatic speech recognition (E2E ASR) models. Recent work on contextual adapters has tackled rare word recognition using user catalogs. This adaptation, however, does not incorporate an important cue, the dialog act, which is available in a multi-turn dialog scenario. In this work, we propose a dialog act guided contextual adapter network. Specifically, it leverages dialog acts to select the most relevant user catalogs and creates queries based on both -- the audio as well as the semantic relationship between the carrier phrase and user catalogs to better guide the contextual biasing. On industrial voice assistant datasets, our model outperforms both the baselines - dialog act encoder-only model, and the contextual adaptation, leading to the most improvement over the no-context model: 58% average relative word error rate reduction (WERR) in the multi-turn dialog scenario, in comparison to the prior-art contextual adapter, which has achieved 39% WERR over the no-context model.【6】 Transformer-based Self-supervised Multimodal Representation Learning for Wearable Emotion Recognition标题：基于变换的可穿戴情感识别的自监督多通道表征学习链接：http://arxiv.org/pdf/2303.17611v1
作者：Yujin Wu,Mohamed Daoudi,Ali Amad机构：University of Lille备注：Accepted IEEE Transactions On Affective Computing摘要：近年来，基于外周生理信号的可穿戴情感识别由于其侵入性小和在现实生活场景中的适用性而引起了广泛的关注。然而，如何有效地融合多模态数据仍然是一个具有挑战性的问题。此外，传统的基于全监督的方法在给定有限的标记数据的情况下遭受过拟合。为了解决上述问题，我们提出了一种新的自监督学习（SSL）框架可穿戴情感识别，其中有效的多模态融合实现与基于时间卷积的模态特定的编码器和基于变换器的共享编码器，捕获模态内和模态间的相关性。广泛的未标记的数据被自动分配标签的五个信号变换，和建议的SSL模型预先训练的信号变换识别为借口的任务，允许提取广义多模态表示的情感相关的下游任务。为了进行评估，首先在大规模自我收集的生理数据集上对所提出的SSL模型进行预训练，随后在三个公共监督的情感识别数据集上冻结或微调所得编码器。最终，我们基于SSL的方法在各种情感分类任务中取得了最先进的结果。同时，该模型被证明是更准确和强大的全监督方法相比，低数据制度。摘要：Recently, wearable emotion recognition based on peripheral physiological signals has drawn massive attention due to its less invasive nature and its applicability in real-life scenarios. However, how to effectively fuse multimodal data remains a challenging problem. Moreover, traditional fully-supervised based approaches suffer from overfitting given limited labeled data. To address the above issues, we propose a novel self-supervised learning (SSL) framework for wearable emotion recognition, where efficient multimodal fusion is realized with temporal convolution-based modality-specific encoders and a transformer-based shared encoder, capturing both intra-modal and inter-modal correlations. Extensive unlabeled data is automatically assigned labels by five signal transforms, and the proposed SSL model is pre-trained with signal transformation recognition as a pretext task, allowing the extraction of generalized multimodal representations for emotion-related downstream tasks. For evaluation, the proposed SSL model was first pre-trained on a large-scale self-collected physiological dataset and the resulting encoder was subsequently frozen or fine-tuned on three public supervised emotion recognition datasets. Ultimately, our SSL-based method achieved state-of-the-art results in various emotion classification tasks. Meanwhile, the proposed model proved to be more accurate and robust compared to fully-supervised methods on low data regimes.机器翻译，仅供参考
每日语音速递