今日论文合集：cs.SD语音4篇，eess.AS音频处理4篇。
cs.SD语音

【1】 Development of a Voice Controlled Robotic Arm

标题：一种声控机械臂的研制链接：http://arxiv.org/pdf/2303.09645v1

作者：Akkas U. Haque,Humayun Kabir,S. C. Banik,M. T. Islam机构：. Speech Application Programming Interface (by Microsoft), DEVELOPMENT OF A VOICE CONTROLLED ROBOTIC ARM, Student, Department of Mechanical Engineering, CUET, Bangladesh, Assistant Professor, Department of Mechanical Engineering, CUET, Bangladesh摘要：本文介绍了一种由中国工程技术大学机电一体化实验室研制的5自由度（DOF）语音控制机械臂。该机械臂通过串行通信（RS-232）与PC连接。用户的语音命令由麦克风采集，并通过Microsoft Visual Studio开发的软件进行处理。然后将特定信号（通过信号处理获得）发送到控制单元。机器人手臂中使用的主控单元是型号为PIC 18f 452的微控制器。然后控制单元根据信号驱动致动器（Hitec HS-422，HS-81），以给出机械臂所需的运动。目前，机械臂可以执行一个设定的动作，如抓和拉，抓，保持和释放，以及一些其他额外的功能，如舞蹈般的运动，并可以根据语音命令转动。摘要：This paper describes a robotic arm with 5 degrees-of-freedom (DOF) which is controlled by human voice and has been developed in the Mechatronics Laboratory, CUET. This robotic arm is interfaced with a PC by serial communication (RS-232). Users' voice command is captured by a microphone, and this voice is processed by software which is made by Microsoft visual studio. Then the specific signal (obtained by signal processing) is sent to control unit. The main control unit that is used in the robotic arm is a microcontroller whose model no. is PIC18f452. Then Control unit drives the actuators, (Hitec HS-422, HS-81) according to the signal or signals to give required motion of the robotic arm. At present the robotic arm can perform a set action like pick & pull, gripping, holding & releasing, and some other extra function like dance-like movement, and can turn according to the voice commands.

【2】 Visual Information Matters for ASR Error Correction标题：视觉信息对ASR纠错的重要性链接：http://arxiv.org/pdf/2303.10160v1

作者：Vanya Bannihatti Kumar,Shanbo Cheng,Ningxin Peng,Yuchen Zhang备注：Accepted at ICASSP 2023机构：ByteDance摘要：为了提高自动语音识别（ASR）的输出效果，ASR纠错技术由于其在使用并行文本数据方面的高效性而得到了广泛的发展。先前的工作主要集中于使用文本或 和语音数据，这在不仅文本和语音信息而且诸如视觉信息的其他模态对于EC都至关重要时阻碍了性能增益。挑战主要有两个方面：一是以往的工作没有强调视觉信息，因此很少进行勘探研究。另一个原因是社区缺乏一个高质量的基准，其中视觉信息对EC模型很重要。为此，本文提出了1）简单而有效的方法，即门控融合和图像字幕作为提示，结合视觉信息来帮助电子商务;2）大规模基准数据集，即Visual-ASR-EC，其中训练数据中的每一项由视觉、语音和文本信息组成，并且测试数据由人类注释者仔细选择，以确保当视觉信息缺失时即使人类也可能出错。实验结果表明，使用字幕作为提示可以有效地利用视觉信息，并且在字错误率（WER）方面比现有方法高出1.2%，这也表明视觉信息在我们提出的Visual-ASR-EC数据集中是至关重要的。摘要：Aiming to improve the Automatic Speech Recognition (ASR) outputs with a post-processing step, ASR error correction (EC) techniques have been widely developed due to their efficiency in using parallel text data. Previous works mainly focus on using text or and speech data, which hinders the performance gain when not only text and speech information, but other modalities, such as visual information are critical for EC. The challenges are mainly two folds: one is that previous work fails to emphasize visual information, thus rare exploration has been studied. The other is that the community lacks a high-quality benchmark where visual information matters for the EC models. Therefore, this paper provides 1) simple yet effective methods, namely gated fusion and image captions as prompts to incorporate visual information to help EC; 2) large-scale benchmark datasets, namely Visual-ASR-EC, where each item in the training data consists of visual, speech, and text information, and the test data are carefully selected by human annotators to ensure that even humans could make mistakes when visual information is missing. Experimental results show that using captions as prompts could effectively use the visual information and surpass state-of-the-art methods by upto 1.2% in Word Error Rate(WER), which also indicates that visual information is critical in our proposed Visual-ASR-EC dataset

【3】 Configurable EBEN: Extreme Bandwidth Extension Network to enhance body-conducted speech capture标题：可配置EBEN：用于增强身体传导语音捕获的极带宽扩展网络链接：http://arxiv.org/pdf/2303.10008v1

作者：Julien Hauret,Thomas Joubaud,Véronique Zimpfer,Éric Bavu机构：HESAM Universit´eParis, net 2nd Joubaud ThomasDepartment of Acoustics and Soldier ProtectionFrench-German Research Institute of Saint-Louis (ISL)Saint-Louis备注：12 pages, 10 figures, 3 tables, submitted to IEEE TASLP摘要：本文介绍了一种可配置版本的极端带宽扩展网络（EBEN），这是一种生成式对抗网络（GAN），旨在改善体导麦克风捕获的音频。我们证明这些麦克风可以显著降低环境噪声。然而，这种对环境噪声的不敏感性是以从设备的佩戴者获取的语音信号的带宽为代价的。因此，所获得的捕获信号需要使用信号增强技术来恢复全带宽语音。EBEN利用原始捕获信号的可配置多频带分解。这种分解允许减少数据时域维度并且更好地控制全频带信号。捕获信号的多频带表示通过类U-Net模型来处理，该模型组合特征和对抗损失以生成增强的语音信号。我们还受益于在建议的可配置鉴别器架构中的这种原始表示。可配置的EBEN方法可以通过允许实时处理的轻量级生成器在合成数据上实现最先进的增强结果。摘要：This paper presents a configurable version of Extreme Bandwidth Extension Network (EBEN), a Generative Adversarial Network (GAN) designed to improve audio captured with body-conduction microphones. We show that these microphones significantly reduce environmental noise. However, this insensitivity to ambient noise is at the expense of the bandwidth of the voice signal acquired from the wearer of the devices. The obtained captured signals therefore require the use of signal enhancement techniques to recover the full-bandwidth speech. EBEN leverages a configurable multiband decomposition of the raw captured signal. This decomposition allows the data time domain dimensions to be reduced and the full band signal to be better controlled. The multiband representation of the captured signal is processed through a U-Net-like model, which combines feature and adversarial losses to generate an enhanced speech signal. We also benefit from this original representation in the proposed configurable discriminator architecture. The configurable EBEN approach can achieve state-of-the-art enhancement results on synthetic data with a lightweight generator that allows real-time processing.

【4】 Magnitude-Corrected and Time-Aligned Interpolation of Head-Related Transfer Functions标题：头部相关传递函数的幅度校正和时间对准内插链接：http://arxiv.org/pdf/2303.09966v1

作者：Johannes M. Arend,Christoph Pörschmann,Stefan Weinzierl,Fabian Brinkmann机构：ID , Audio Communication Group, Technical University of Berlin, Berlin , Germany, Institute of Computer and Communication Technology, TH Köln - University of Applied Sciences, Cologne , Germany摘要：头部相关传递函数（HRTF）是虚拟声学现实的关键，因为它包含了在三维空间中定位声源的所有线索。声学测量是获得高质量HRTF的一种方法。为了减少测量系统的测量时间、成本和复杂度，一种有前途的方法是在稀疏采样网格上仅捕获少数HRTF，然后通过插值将其上采样到密集HRTF集合。然而，HRTF插值具有挑战性，因为源位置的微小变化可能导致HRTF相位和幅度响应的显著变化。先前的研究通过在预处理中对HRTF进行时间对齐，极大地改进了插值，但是幅度插值误差，特别是在对侧区域，仍然是一个问题。在时间对齐方法的基础上，我们提出了一种基于频率平滑HRTF表示的插值后幅度校正方法。使用HUTUBS数据库的所有96个单独的模拟HRTF集，我们表明，与仅应用时间对齐的最先进插值方法相比，震级校正显著降低了插值误差。我们的分析表明，当对非常稀疏的HRTF集进行上采样时，当在所有方向上进行平均时，临界较高频率范围中的受试者平均幅度误差最多低1.5 dB，在对侧区域中甚至最多低4 dB。结果，显著改善了上采样HRTF中的耳间电平差。因此，所提出的算法具有进一步减少感知透明插值所需的HRTF的最小数目的潜力。摘要：Head-related transfer functions (HRTFs) are essential for virtual acoustic realities, as they contain all cues for localizing sound sources in three-dimensional space. Acoustic measurements are one way to obtain high-quality HRTFs. To reduce measurement time, cost, and complexity of measurement systems, a promising approach is to capture only a few HRTFs on a sparse sampling grid and then upsample them to a dense HRTF set by interpolation. However, HRTF interpolation is challenging because small changes in source position can result in significant changes in the HRTF phase and magnitude response. Previous studies greatly improved the interpolation by time-aligning the HRTFs in preprocessing, but magnitude interpolation errors, especially in contralateral regions, remain a problem. Building upon the time-alignment approaches, we propose an additional post-interpolation magnitude correction derived from a frequency-smoothed HRTF representation. Employing all 96 individual simulated HRTF sets of the HUTUBS database, we show that the magnitude correction significantly reduces interpolation errors compared to state-of-the-art interpolation methods applying only time alignment. Our analysis shows that when upsampling very sparse HRTF sets, the subject-averaged magnitude error in the critical higher frequency range is up to 1.5 dB lower when averaged over all directions and even up to 4 dB lower in the contralateral region. As a result, the interaural level differences in the upsampled HRTFs are considerably improved. The proposed algorithm thus has the potential to further reduce the minimum number of HRTFs required for perceptually transparent interpolation.
eess.AS音频处理

【1】 Visual Information Matters for ASR Error Correction标题：视觉信息对ASR纠错的重要性链接：http://arxiv.org/pdf/2303.10160v1

*与cs.SD语音【2】为同一篇

作者：Vanya Bannihatti Kumar,Shanbo Cheng,Ningxin Peng,Yuchen Zhang备注：Accepted at ICASSP 2023机构：ByteDance摘要：为了提高自动语音识别（ASR）的输出效果，ASR纠错技术由于其在使用并行文本数据方面的高效性而得到了广泛的发展。先前的工作主要集中于使用文本或 和语音数据，这在不仅文本和语音信息而且诸如视觉信息的其他模态对于EC都至关重要时阻碍了性能增益。挑战主要有两个方面：一是以往的工作没有强调视觉信息，因此很少进行勘探研究。另一个原因是社区缺乏一个高质量的基准，其中视觉信息对EC模型很重要。为此，本文提出了1）简单而有效的方法，即门控融合和图像字幕作为提示，结合视觉信息来帮助电子商务;2）大规模基准数据集，即Visual-ASR-EC，其中训练数据中的每一项由视觉、语音和文本信息组成，并且测试数据由人类注释者仔细选择，以确保当视觉信息缺失时即使人类也可能出错。实验结果表明，使用字幕作为提示可以有效地利用视觉信息，并且在字错误率（WER）方面比现有方法高出1.2%，这也表明视觉信息在我们提出的Visual-ASR-EC数据集中是至关重要的。摘要：Aiming to improve the Automatic Speech Recognition (ASR) outputs with a post-processing step, ASR error correction (EC) techniques have been widely developed due to their efficiency in using parallel text data. Previous works mainly focus on using text or and speech data, which hinders the performance gain when not only text and speech information, but other modalities, such as visual information are critical for EC. The challenges are mainly two folds: one is that previous work fails to emphasize visual information, thus rare exploration has been studied. The other is that the community lacks a high-quality benchmark where visual information matters for the EC models. Therefore, this paper provides 1) simple yet effective methods, namely gated fusion and image captions as prompts to incorporate visual information to help EC; 2) large-scale benchmark datasets, namely Visual-ASR-EC, where each item in the training data consists of visual, speech, and text information, and the test data are carefully selected by human annotators to ensure that even humans could make mistakes when visual information is missing. Experimental results show that using captions as prompts could effectively use the visual information and surpass state-of-the-art methods by upto 1.2% in Word Error Rate(WER), which also indicates that visual information is critical in our proposed Visual-ASR-EC dataset

【2】 Configurable EBEN: Extreme Bandwidth Extension Network to enhance body-conducted speech capture标题：可配置EBEN：用于增强身体传导语音捕获的极带宽扩展网络链接：http://arxiv.org/pdf/2303.10008v1

*与cs.SD语音【3】为同一篇

作者：Julien Hauret,Thomas Joubaud,Véronique Zimpfer,Éric Bavu机构：HESAM Universit´eParis, net 2nd Joubaud ThomasDepartment of Acoustics and Soldier ProtectionFrench-German Research Institute of Saint-Louis (ISL)Saint-Louis备注：12 pages, 10 figures, 3 tables, submitted to IEEE TASLP摘要：本文介绍了一种可配置版本的极端带宽扩展网络（EBEN），这是一种生成式对抗网络（GAN），旨在改善体导麦克风捕获的音频。我们证明这些麦克风可以显著降低环境噪声。然而，这种对环境噪声的不敏感性是以从设备的佩戴者获取的语音信号的带宽为代价的。因此，所获得的捕获信号需要使用信号增强技术来恢复全带宽语音。EBEN利用原始捕获信号的可配置多频带分解。这种分解允许减少数据时域维度并且更好地控制全频带信号。捕获信号的多频带表示通过类U-Net模型来处理，该模型组合特征和对抗损失以生成增强的语音信号。我们还受益于在建议的可配置鉴别器架构中的这种原始表示。可配置的EBEN方法可以通过允许实时处理的轻量级生成器在合成数据上实现最先进的增强结果。摘要：This paper presents a configurable version of Extreme Bandwidth Extension Network (EBEN), a Generative Adversarial Network (GAN) designed to improve audio captured with body-conduction microphones. We show that these microphones significantly reduce environmental noise. However, this insensitivity to ambient noise is at the expense of the bandwidth of the voice signal acquired from the wearer of the devices. The obtained captured signals therefore require the use of signal enhancement techniques to recover the full-bandwidth speech. EBEN leverages a configurable multiband decomposition of the raw captured signal. This decomposition allows the data time domain dimensions to be reduced and the full band signal to be better controlled. The multiband representation of the captured signal is processed through a U-Net-like model, which combines feature and adversarial losses to generate an enhanced speech signal. We also benefit from this original representation in the proposed configurable discriminator architecture. The configurable EBEN approach can achieve state-of-the-art enhancement results on synthetic data with a lightweight generator that allows real-time processing.

【3】 Magnitude-Corrected and Time-Aligned Interpolation of Head-Related Transfer Functions标题：头部相关传递函数的幅度校正和时间对准内插链接：http://arxiv.org/pdf/2303.09966v1

*与cs.SD语音【4】为同一篇

作者：Johannes M. Arend,Christoph Pörschmann,Stefan Weinzierl,Fabian Brinkmann机构：ID , Audio Communication Group, Technical University of Berlin, Berlin , Germany, Institute of Computer and Communication Technology, TH Köln - University of Applied Sciences, Cologne , Germany摘要：头部相关传递函数（HRTF）是虚拟声学现实的关键，因为它包含了在三维空间中定位声源的所有线索。声学测量是获得高质量HRTF的一种方法。为了减少测量系统的测量时间、成本和复杂度，一种有前途的方法是在稀疏采样网格上仅捕获少数HRTF，然后通过插值将其上采样到密集HRTF集合。然而，HRTF插值具有挑战性，因为源位置的微小变化可能导致HRTF相位和幅度响应的显著变化。先前的研究通过在预处理中对HRTF进行时间对齐，极大地改进了插值，但是幅度插值误差，特别是在对侧区域，仍然是一个问题。在时间对齐方法的基础上，我们提出了一种基于频率平滑HRTF表示的插值后幅度校正方法。使用HUTUBS数据库的所有96个单独的模拟HRTF集，我们表明，与仅应用时间对齐的最先进插值方法相比，震级校正显著降低了插值误差。我们的分析表明，当对非常稀疏的HRTF集进行上采样时，当在所有方向上进行平均时，临界较高频率范围中的受试者平均幅度误差最多低1.5 dB，在对侧区域中甚至最多低4 dB。结果，显著改善了上采样HRTF中的耳间电平差。因此，所提出的算法具有进一步减少感知透明插值所需的HRTF的最小数目的潜力。摘要：Head-related transfer functions (HRTFs) are essential for virtual acoustic realities, as they contain all cues for localizing sound sources in three-dimensional space. Acoustic measurements are one way to obtain high-quality HRTFs. To reduce measurement time, cost, and complexity of measurement systems, a promising approach is to capture only a few HRTFs on a sparse sampling grid and then upsample them to a dense HRTF set by interpolation. However, HRTF interpolation is challenging because small changes in source position can result in significant changes in the HRTF phase and magnitude response. Previous studies greatly improved the interpolation by time-aligning the HRTFs in preprocessing, but magnitude interpolation errors, especially in contralateral regions, remain a problem. Building upon the time-alignment approaches, we propose an additional post-interpolation magnitude correction derived from a frequency-smoothed HRTF representation. Employing all 96 individual simulated HRTF sets of the HUTUBS database, we show that the magnitude correction significantly reduces interpolation errors compared to state-of-the-art interpolation methods applying only time alignment. Our analysis shows that when upsampling very sparse HRTF sets, the subject-averaged magnitude error in the critical higher frequency range is up to 1.5 dB lower when averaged over all directions and even up to 4 dB lower in the contralateral region. As a result, the interaural level differences in the upsampled HRTFs are considerably improved. The proposed algorithm thus has the potential to further reduce the minimum number of HRTFs required for perceptually transparent interpolation.

【4】 Development of a Voice Controlled Robotic Arm

标题：一种声控机械臂的研制链接：http://arxiv.org/pdf/2303.09645v1

*与cs.SD语音【1】为同一篇

作者：Akkas U. Haque,Humayun Kabir,S. C. Banik,M. T. Islam机构：. Speech Application Programming Interface (by Microsoft), DEVELOPMENT OF A VOICE CONTROLLED ROBOTIC ARM, Student, Department of Mechanical Engineering, CUET, Bangladesh, Assistant Professor, Department of Mechanical Engineering, CUET, Bangladesh摘要：本文介绍了一种由中国工程技术大学机电一体化实验室研制的5自由度（DOF）语音控制机械臂。该机械臂通过串行通信（RS-232）与PC连接。用户的语音命令由麦克风采集，并通过Microsoft Visual Studio开发的软件进行处理。然后将特定信号（通过信号处理获得）发送到控制单元。机器人手臂中使用的主控单元是型号为PIC 18f 452的微控制器。然后控制单元根据信号驱动致动器（Hitec HS-422，HS-81），以给出机械臂所需的运动。目前，机械臂可以执行一个设定的动作，如抓和拉，抓，保持和释放，以及一些其他额外的功能，如舞蹈般的运动，并可以根据语音命令转动。摘要：This paper describes a robotic arm with 5 degrees-of-freedom (DOF) which is controlled by human voice and has been developed in the Mechatronics Laboratory, CUET. This robotic arm is interfaced with a PC by serial communication (RS-232). Users' voice command is captured by a microphone, and this voice is processed by software which is made by Microsoft visual studio. Then the specific signal (obtained by signal processing) is sent to control unit. The main control unit that is used in the robotic arm is a microcontroller whose model no. is PIC18f452. Then Control unit drives the actuators, (Hitec HS-422, HS-81) according to the signal or signals to give required motion of the robotic arm. At present the robotic arm can perform a set action like pick & pull, gripping, holding & releasing, and some other extra function like dance-like movement, and can turn according to the voice commands.

机器翻译，仅供参考