In speech enhancement, an end-to-end deep neural network converts a noisy speech signal to a clean speech directly in time domain without time-frequency transformation or mask estimation. However, aggregating contextual information from a high-resolution time domain signal with an affordable model complexity still remains challenging. In this paper, we propose a densely connected convolutional and recurrent network (DCCRN), a hybrid architecture, to enable dual-staged temporal context aggregation. With the dense connectivity and cross-component identical shortcut, DCCRN consistently outperforms competing convolutional baselines with an average STOI improvement of 0.23 and PESQ of 1.38 at three SNR levels. The proposed method is computationally efficient with only 1.38 million parameters. The generalizability performance on the unseen noise types is still decent considering its low complexity, although it is relatively weaker comparing to Wave-U-Net with 7.25 times more parameters.
Rethinking Speech Representation Aggregation in Speech Enhancement: A Phonetic Mutual Information Perspective
Department of Intelligence and Information, Seoul National University, Republic of Korea（智能与信息系，首尔国立大学，韩国）；Artificial Intelligence Institute, Seoul National University, Republic of Korea（人工智能研究所，首尔国立大学，韩国）
In this paper, we propose a multi-channel network for simultaneous speech dereverberation, enhancement and separation (DESNet). To enable gradient propagation and joint optimization, we adopt the attentional selection mechanism of the multi-channel features, which is originally proposed in end-to-end unmixing, fixed-beamforming and extraction (E2E-UFE) structure. Furthermore, the novel deep complex convolutional recurrent network (DCCRN) is used as the structure of the speech unmixing and the neural network based weighted prediction error (WPE) is cascaded beforehand for speech dereverberation…
Institute for Integrated and Intelligent Systems, Griffith University, Australia（格里菲斯大学综合智能系统研究所）；School of Artificial Intelligence, Xidian University, China（西安电子科技大学人工智能学院）
Convolutional neural networks (CNNs) with residual links (ResNets) and causal dilated convolutional units have been the network of choice for deep learning approaches to speech enhancement. While residual links improve gradient flow during training, feature diminution of shallow layer outputs can occur due to repetitive summations with deeper layer outputs. One strategy to improve feature re-usage is to fuse both ResNets and densely connected CNNs (DenseNets). DenseNets, however, over-allocate parameters for feature re-usage. Motivated by this, we propose the residual-dense lattice network (RD…
In automatic speech recognition, often little training data is available for specific challenging tasks, but training of state-of-the-art automatic speech recognition systems requires large amounts of annotated speech. To address this issue, we propose a two-staged approach to acoustic modeling that combines noise and reverberation data augmentation with transfer learning to robustly address challenges such as difficult acoustic recording conditions, spontaneous speech, and speech of elderly people. We evaluate our approach using the example of German oral history interviews, where a relative …
In this work, we propose a new model called triple-path attentive recurrent network (TPARN) for multichannel speech enhancement in the time domain. TPARN extends a single-channel dual-path network to a multichannel network by adding a third path along the spatial dimension. First, TPARN processes speech signals from all channels independently using a dual-path attentive recurrent network (ARN), which is a recurrent neural network (RNN) augmented with self-attention. Next, an ARN is introduced along the spatial dimension for spatial context aggregation. TPARN is designed as a multiple-input and…
Speech enhancement in the time domain is becoming increasingly popular in recent years, due to its capability to jointly enhance both the magnitude and the phase of speech. In this work, we propose a dense convolutional network (DCN) with self-attention for speech enhancement in the time domain. DCN is an encoder and decoder based architecture with skip connections. Each layer in the encoder and the decoder comprises a dense block and an attention module. Dense blocks and attention modules help in feature extraction using a combination of feature reuse, increased network depth, and maximum con…
We present an end-to-end deep learning approach to denoising speech signals by processing the raw waveform directly. Given input audio containing speech corrupted by an additive background signal, the system aims to produce a processed signal that contains only the speech content. Recent approaches have shown promising results using various deep network architectures. In this paper, we propose to train a fully-convolutional context aggregation network using a deep feature loss. That loss is based on comparing the internal feature activations in a different network, trained for acoustic environ…
VoiceWeaver: Staged Learning of Structured Controls for Expressive Speech and Sound-Event Generation
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）
Georgia Institute of Technology（佐治亚理工学院）；Universidad San Pablo-CEU（圣巴布洛CEU大学）
Distributed automatic speech recognition (ASR) requires to aggregate outputs of distributed deep neural network (DNN)-based models. This work studies the use of submodular functions to design a rank aggregation on score-based permutations, which can be used for distributed ASR systems in both supervised and unsupervised modes. Specifically, we compose an aggregation rank function based on the Lovasz Bregman divergence for setting up linear structured convex and nested structured concave functions. The algorithm is based on stochastic gradient descent (SGD) and can obtain well-trained aggregati…
Learning emotion embedding from reference audio is a straightforward approach for multi-emotion speech synthesis in encoder-decoder systems. But how to get better emotion embedding and how to inject it into TTS acoustic model more effectively are still under investigation. In this paper, we propose an innovative constraint to help VAE extract emotion embedding with better cluster cohesion. Besides, the obtained emotion embedding is used as query to aggregate latent representations of all encoder layers via attention. Moreover, the queries from encoder layers themselves are also helpful. Experi…
Differential privacy (DP) is one data protection avenue to safeguard user information used for training deep models by imposing noisy distortion on privacy data. Such a noise perturbation often results in a severe performance degradation in automatic speech recognition (ASR) in order to meet a privacy budget $\varepsilon$. Private aggregation of teacher ensemble (PATE) utilizes ensemble probabilities to improve ASR accuracy when dealing with the noise effects controlled by small values of $\varepsilon$. We extend PATE learning to work with dynamic patterns, namely speech utterances, and perfor…
University of Cambridge（剑桥大学）；Avignon University（阿维尼翁大学）；Telecom Paris（巴黎电信）；University of Oxford（牛津大学）；Adap GmbH（Adap有限责任公司）
Training Automatic Speech Recognition (ASR) models under federated learning (FL) settings has attracted a lot of attention recently. However, the FL scenarios often presented in the literature are artificial and fail to capture the complexity of real FL systems. In this paper, we construct a challenging and realistic ASR federated experimental setup consisting of clients with heterogeneous data distributions using the French and Italian sets of the CommonVoice dataset, a large heterogeneous dataset containing thousands of different speakers, acoustic environments and noises. We present the fir…
Attention-based seq2seq text-to-speech systems, especially those use self-attention networks (SAN), have achieved state-of-art performance. But an expressive corpus with rich prosody is still challenging to model as 1) prosodic aspects, which span across different sentential granularities and mainly determine acoustic expressiveness, are difficult to quantize and label and 2) the current seq2seq framework extracts prosodic information solely from a text encoder, which is easily collapsed to an averaged expression for expressive contents. In this paper, we propose a context extractor, which is …
Unimodal Aggregation for CTC-based Speech Recognition
Zhejiang University（浙江大学）；School of Engineering, Westlake University（西湖大学工程学院）；Institute of Advanced Technology, Westlake Institute for Advanced Study（西湖高等研究院先进技术研究所）
Georgia Institute of Technology（佐治亚理工学院）；University of Enna（恩纳大学）；NTNU（挪威科技大学）
We propose using an adversarial autoencoder (AAE) to replace generative adversarial network (GAN) in the private aggregation of teacher ensembles (PATE), a solution for ensuring differential privacy in speech applications. The AAE architecture allows us to obtain good synthetic speech leveraging upon a discriminative training of latent vectors. Such synthetic speech is used to build a privacy-preserving classifier when non-sensitive data is not sufficiently available in the public domain. This classifier follows the PATE scheme that uses an ensemble of noisy outputs to label the synthetic samp…
In this paper we extend the x-vector framework for the task of speaker's age estimation and gender classification. In particular, we replace the baseline multilayer-TDNN architecture with QuartzNet, a convolutional architecture that has gained success in the field of speech recognition. We further propose a two-staged transfer learning scheme, utilizing large scale speech datasets: VoxCeleb and Common Voice, and usage of multitask learning to allow for joint age estimation and gender classification with a single system. We train and evaluate the performance on the TIMIT dataset. The proposed t…
Recently proposed speech recognition systems are designed to predict using representations generated by their top layers, employing greedy decoding which isolates each timestep from the rest of the sequence. Aiming for improved performance, a beam search algorithm is frequently utilized and a language model is incorporated to assist with ranking the top candidates. In this work, we experiment with several speech recognition models and find that logits predicted using the top layers may hamper beam search from achieving optimal results. Specifically, we show that fined-tuned Wav2Vec 2.0 and HuB…
Speaker-independent speech separation has achieved remarkable performance in recent years with the development of deep neural network (DNN). Various network architectures, from traditional convolutional neural network (CNN) and recurrent neural network (RNN) to advanced transformer, have been designed sophistically to improve separation performance. However, the state-of-the-art models usually suffer from several flaws related to the computation, such as large model size, huge memory consumption and computational complexity. To find the balance between the performance and computational efficie…
Robust Rank Aggregation for Multimodal Speech-Based Alzheimer's Disease Detection
The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Center for Language, Intelligence and Machines (LIMA), Shenzhen Loop Area Institute (SLAI)（深圳环区研究所（SLAI）语言、智能与机器中心（LIMA））
Progressive Facial Granularity Aggregation with Bilateral Attribute-based Enhancement for Face-to-Speech Synthesis
Graduate School of Artificial Intelligence, POSTECH, Republic of Korea（人工智能研究生院，POSTECH，韩国）；Department of Computer Science and Engineering, POSTECH, Republic of Korea（计算机科学与工程系，POSTECH，韩国）
Domain-specific data is the crux of the successful transfer of machine learning systems from benchmarks to real life. In simple problems such as image classification, crowdsourcing has become one of the standard tools for cheap and time-efficient data collection: thanks in large part to advances in research on aggregation methods. However, the applicability of crowdsourcing to more complex tasks (e.g., speech recognition) remains limited due to the lack of principled aggregation methods for these modalities. The main obstacle towards designing aggregation methods for more advanced applications…
We introduce the novel concept of anti-transfer learning for speech processing with convolutional neural networks. While transfer learning assumes that the learning process for a target task will benefit from re-using representations learned for another task, anti-transfer avoids the learning of representations that have been learned for an orthogonal task, i.e., one that is not relevant and potentially misleading for the target task, such as speaker identity for speech recognition or speech content for emotion recognition. In anti-transfer learning, we penalize similarity between activations …
Seoul National University（首尔大学）；Advanced Institutes of Convergence Technology（先进融合技术研究院）
In recent years, neural network approaches have shown superior performance to conventional hand-made features in numerous application areas. In particular, convolutional neural networks (ConvNets) exploit spatially local correlations across input data to improve the performance of audio processing tasks, such as speech recognition, musical chord recognition, and onset detection. Here we apply ConvNet to acoustic scene classification, and show that the error rate can be further decreased by using delta features in the frequency domain. We propose a multiple-width frequency-delta (MWFD) data aug…
