Sitemap A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Page not found. Your pixels are in another canvas.
This is a page not in th emain menu
Posts less than 1 minute read
Published: November 14, 2024
恭喜!
项目“乘用车座舱声景设计与智能控制关键技术”荣获2024中国汽车工程学会科学技术奖,科学技术进步奖,二等奖. less than 1 minute read
Published: October 23, 2024
恭喜!
论文“Spectral network based on lattice convolution and adversarial training for noise-robust speech super-resolution”被 The Journal of the Acoustical Society of America (JASA)接收. less than 1 minute read
Published: September 29, 2024
恭喜!
论文“SDNet: Noise-Robust Bandwidth Extension under Flexible Sampling Rates”被 Asia Pacific Signal and Information Processing Association (APSIPA), 2024接收。 论文“Diverse Time-Frequency Attention Neural Network for Acoustic Echo Cancellation”被 Asia Pacific Signal and Information Processing Association (APSIPA), 2024接收。 less than 1 minute read
Published: September 18, 2024
恭喜论文“Cross Domain Optimization for Speech Enhancement: Parallel or Cascade?”被 IEEE/ACM Transactions on Audio, Speech and Language Processing接收.
less than 1 minute read
Published: August 14, 2024
恭喜团队学生论文‘On the Analysis of GAN-based Image-to-Image Translation with Gaussian Noise Injection’被机器学习顶会ICLR 2024接收。
less than 1 minute read
Published: July 23, 2024
恭喜团队提出的音视频融合检测与定位框架,在DCASE 2024 Task 3.A&B 音视频融合声事件检测与定位均获得全球第四名。
less than 1 minute read
Published: April 14, 2024
团队论文《Towards optimized multi-channel modulo-ADCS: moduli selection strategies and bit depth analysis》成功发表在《ICASSP 2024》 团队论文《Understanding Gaussian Noise Mismatch: A Hellinger Distance Approach》成功发表在《ICASSP 2024》 less than 1 minute read
Published: August 14, 2023
团队研究生邓卓文的信源分离论文《Time-frequency Domain Filter-and-sum Network for Multi-channel Speech Separation》成功发表在《INTERSPEECH 2023》 团队本科生帅辰昊和石超华语音超分辨论文《mdctGAN: Taming transformer-based GAN for speech super-resolution with Modified DCT spectra》成功发表在《INTERSPEECH 2023》 less than 1 minute read
Published: January 01, 2023
恭喜团队学生在DCASE 2023 Task 3.A 音视频融合声事件检测与定位获得全球第二名。
less than 1 minute read
Published: November 23, 2022
恭喜团队L3DAS22: Machine Learning for 3D Audio Signal Processing - 3D Sound Event Localization and Detection在新加坡2022 ICASSP获得全球第二名。
less than 1 minute read
Published: October 23, 2022
恭喜团队’踏“铁”寻声–高铁底盘稳定速检供应商’,在2022年互联网+大学生创新创业大赛获得重庆市金奖。
less than 1 minute read
Published: August 14, 2022
恭喜团队在2022 DCASE Task 2 无监督机器状态监测中获得全球第一名。该项目主要是完成机器状态的监测,例如发动机,轴承等机器工作是否存在异常,免去人工监测成本。该项成果在工业界有着广泛的应用。
less than 1 minute read
Published: October 15, 2021
恭喜团队 “Timbre” - 一款基于计算机听觉技术的智能安全监测机器人,获得2021全国互联网+大学生创新创业大赛全国铜奖。
less than 1 minute read
Published: August 14, 2021
恭喜团队的’Windblade listener - an intelligent approach for wind turbine blade monitoring using sound’在2021全国互联网+大学生创新创业大赛获得全国银奖。该项目主要是完成风力发电机的扇叶转动状态的评估,判断扇叶是否存在异常,例如结冰,裂缝等,极大的提高自动化水平。
less than 1 minute read
Published: July 23, 2020
恭喜团队Beamforming Videos在2020西班牙ICASSP获得Global Top 10.
less than 1 minute read
Published: October 15, 2019
恭喜团队的’The intelligent wheelchair based on GIS and MIC array speech recognition’在2019全国互联网+大学生创新创业大赛获得全国银奖。
portfolio
Implementation of RCPO into stable-baselines3 PPO.
publications talks
回声消除(Acoustic echo cancellation, AEC)是利用远端信号作为参考信号,完成近端信号中的回声去除,完成清晰的通话。
主动降噪(Active noise control, ANC)是利用声波的反向特性,完成噪声的主动抵消,以打造静谧空间。
利用transformer结构的encoder-decoder方案,可以实现轻松的转录。
电动汽车在纯电动模式下以低速行驶时,由于噪声低使得道路行人不容易察觉到车辆的接近,发生交通事故。为此,低速行使状态下,电动车应该主动发声,提示行人。
大模型的迅速发展,为其他下游任务(downstream tasks)的应用提供了更多可能。
音质修复包含很多方面,例如噪声去除,啸叫抑制,语音超分辨等,完成语音质量的提升。
声事件检测(Acoustic(sound) event detection, A(S)ED)是对声音中所包含的声音种类进行判别。
信源分离(Source separation, SS)是鸡尾酒会的代名词,是指从多个说话人混合的信号中提取每个说话人。
通过明区和暗区的声场的控制,实现不同声场的空间分配。
波束形成(Beamforming, BF)是指利用麦克风阵列,完成指向性操作,主瓣内的信号可以无损的通过,旁瓣的信号可以得到有效的抑制。 teaching