This monograph provides an overview of general deep learning methodology and its applications to a variety of signal and information processing tasks. The application areas are chosen with the following three criteria in mind: (1) expertise or knowledge of the authors; (2) the application areas that have already been transformed by the successful use of deep learning technology, such as speech recognition and computer vision; and (3) the application areas that have the potential to be impacted significantly by deep learning and that have been experiencing research growth, including natural language and text processing, information retrieval, and multimodal information processing empowered by multi-task deep learning.

/pdf/deep-learning-methods-and-applications-5cyi610sjz.pdf

Deep Learning: Methods and Applications

In recent years, deep neural networks have been successful in both industry and academia, especially for computer vision tasks. The great success of deep learning is mainly due to its scalability to encode large-scale data and to maneuver billions of model parameters. However, it is a challenge to deploy these cumbersome deep models on devices with limited resources, e.g., mobile phones and embedded devices, not only because of the high computational complexity but also the large storage requirements. To this end, a variety of model compression and acceleration techniques have been developed. As a representative type of model compression and acceleration, knowledge distillation effectively learns a small student model from a large teacher model. It has received rapid increasing attention from the community. This paper provides a comprehensive survey of knowledge distillation from the perspectives of knowledge categories, training schemes, teacher-student architecture, distillation algorithms, performance comparison and applications. Furthermore, challenges in knowledge distillation are briefly reviewed and comments on future research are discussed and forwarded.

Knowledge Distillation: A Survey

The CHiME challenge series aims to advance far field speech recognition technology by promoting research at the interface of signal processing and automatic speech recognition. This paper presents the design and outcomes of the 3rd CHiME Challenge, which targets the performance of automatic speech recognition in a real-world, commercially-motivated scenario: a person talking to a tablet device that has been fitted with a six-channel microphone array. The paper describes the data collection, the task definition and the baseline systems for data simulation, enhancement and recognition. The paper then presents an overview of the 26 systems that were submitted to the challenge focusing on the strategies that proved to be most successful relative to the MVDR array processing and DNN acoustic modeling reference system. Challenge findings related to the role of simulated data in system training and evaluation are discussed.

/pdf/the-third-chime-speech-separation-and-recognition-challenge-4108e6hzs1.pdf

The third ‘CHiME’ speech separation and recognition challenge: Dataset, task and baselines

http://cs.uef.fi/pages/tkinnu/webpage/pdf/spoofingsurvey2014_SPCOM.pdf

Spoofing and countermeasures for speaker verification

Where you can find the array signal processing easily? Is it in the book store? On-line book store? are you sure? Keep in mind that you will find the book in this site. This book is very referred for you because it gives not only the experience but also lesson. The lessons are very valuable to serve for you, that's not about who are reading this array signal processing book. It is about this book that will give wellness for all people from many societies.

Array signal processing

This paper presents a voice conversion technique using Deep Belief Nets (DBNs) to build high-order eigen spaces of the source/target speakers, where it is easier to convert the source speech to the target speech than in the traditional cepstrum space. DBNs have a deep architecture that automatically discovers abstractions to maximally express the original input features. If we train the DBNs using only the speech of an individual speaker, it can be considered that there is less phonological information and relatively more speaker individuality in the output features at the highest layer. Training the DBNs for a source speaker and a target speaker, we can then connect and convert the speaker individuality abstractions using Neural Networks (NNs). The converted abstraction of the source speaker is then brought back to the cepstrum space using an inverse process of the DBNs of the target speaker. We conducted speakervoice conversion experiments and confirmed the efficacy of our method with respect to subjective and objective criteria, comparing it with the conventional Gaussian Mixture Model-based method.

/pdf/voice-conversion-in-high-order-eigen-space-using-deep-belief-2og9yjdeao.pdf

Voice Conversion in High-order Eigen Space Using Deep Belief Nets

This paper presents a voice conversion (VC) technique for noisy environments, where parallel exemplars are introduced to encode the source speech signal and synthesize the target speech signal. The parallel exemplars (dictionary) consist of the source exemplars and target exemplars, having the same texts uttered by the source and target speakers. The input source signal is decomposed into the source exemplars, noise exemplars obtained from the input signal, and their weights (activities). Then, by using the weights of the source exemplars, the converted signal is constructed from the target exemplars. We carried out speaker conversion tasks using clean speech data and noise-added speech data. The effectiveness of this method was confirmed by comparing its effectiveness with that of a conventional Gaussian Mixture Model (GMM)-based method.

/pdf/exemplar-based-voice-conversion-in-noisy-environment-3s8fk3380l.pdf

Exemplar-based voice conversion in noisy environment

We propose Gaussian Mixture Model (GMM)-based emotional voice conversion using spectrum and prosody features. In recent years, speech recognition and synthesis techniques have been developed, and an emotional voice conversion technique is required for synthesizing more exp ressive voices. The common emotional conversion was based on transformation of neutral prosody to emotional prosody by using huge speech corpus. In this paper, we convert a neutral voice to an emot ional vo ice using GMMs. GMM-based spectrum conversion is widely used to modify non linguistic informat ion such as voice characteristics while keep ing linguistic information unchanged. Because the conventional method converts either prosody or voice quality (spectrum), som e emot ions are not converted well. In our method, both prosody and voice quality are used for converting a neutral voice to an emotional voice, and it is able to obtain more expressive voices in comparison with conventional methods, such as prosody or spectrum conversion.

/pdf/gmm-based-emotional-voice-conversion-using-spectrum-and-1362vc7lv9.pdf

GMM-Based Emotional Voice Conversion Using Spectrum and Prosody Features

End-to-end acoustic models, such as connectionist temporal classification (CTC) and the attention model, have been studied, and their speech recognition accuracies come close to those of conventional deep neural network (DNN)-hidden Markov models. However, most high-performance end-to-end models are not suitable for real-time (streaming) speech recognition because they are based on bidirectional recurrent neural networks (RNNs). In this study, to improve the performance of unidirectional RNN-based CTC, which is suitable for real-time processing, we investigate the knowledge distillation (KD)-based model compression method for training a CTC acoustic model. we evaluate a frame-level KD method and a sequence-level KD method for CTC model. The speech recognition experiments on Wall Street Journal tasks demonstrate that, the frame-level KD worsens the WERs ofunidirectional CTC model, whereas sequence-level KD can improve the WERs of the model.

An Investigation of a Knowledge Distillation Method for CTC Acoustic Models

Objective: The prosody of children with autism spectrum disorder (ASD) has several abnormal features. We assessed the speech tone of children with ASD and of children with typical development (TD) by using a new quantitative acoustic analysis. Methods : Our study participants consisted of 63 children (26 with ASD and 37 with TD). The participants were divided into 4 groups based on their developmental features and age. We assessed the variety of the fundamental frequency (F0) pattern quantitatively, using pitch coefficient of variation (CV), considering the different F0 mean for each word. Results: (1) No significant difference was observed between the ASD and TD group at pre-school age. However, the TD group exhibited significantly greater pitch CV than the ASD group at school age. (2) In pitch CV, range and standard deviation of the whole speech of each participant, no significant differences were observed between the type of participants and age. (3) No significant correlation was found between the pitch CV of each word and the Japanese Autism Screening Questionnaire total score, or between the pitch CV of each word and the intelligence quotient levels in the ASD group. A significant correlation was observed between the pitch CV of each word and social reciprocal interaction score. Conclusions: We assessed the speech tone of children with ASD by using a new quantitative method. Monotonous speech in school-aged children with ASD was detected. The extent of monotonous speech was related to the extent of social reciprocal interaction in children with ASD.

/pdf/speech-intonation-in-children-with-autism-spectrum-disorder-1x3d656n5i.pdf

Ryoichi Takashima

Papers

Voice Conversion in High-order Eigen Space Using Deep Belief Nets

Exemplar-based voice conversion in noisy environment

GMM-Based Emotional Voice Conversion Using Spectrum and Prosody Features

An Investigation of a Knowledge Distillation Method for CTC Acoustic Models

Speech intonation in children with autism spectrum disorder