We propose a tensor-to-vector regression approach to multi-channel speech enhancement in order to address the issue of input size explosion and hidden-layer size expansion. The key idea is to cast the conventional deep neural network (DNN) based vector-to-vector regression formulation under a tensor-train network (TTN) framework. TTN is a recently emerged solution for compact representation of deep models with fully connected hidden layers. Thus TTN maintains DNN's expressive power yet involves a much smaller amount of trainable parameters. Furthermore, TTN can handle a multi-dimensional tensor input by design, which exactly matches the desired setting in multi-channel speech enhancement. We first provide a theoretical extension from DNN to TTN based regression. Next, we show that TTN can attain speech enhancement quality comparable with that for DNN but with much fewer parameters, e.g., a reduction from 27 million to only 5 million parameters is observed in a single-channel scenario. TTN also improves PESQ over DNN from 2.86 to 2.96 by slightly increasing the number of trainable parameters. Finally, in 8-channel conditions, a PESQ of 3.12 is achieved using 20 million parameters for TTN, whereas a DNN with 68 million parameters can only attain a PESQ of 3.06. Our implementation is available online https://github.com/uwjunqi/Tensor-Train-Neural-Network.
Speech emotion recognition~(SER) refers to the technique of inferring the emotional state of an individual from speech signals. SERs continue to garner interest due to their wide applicability. Although the domain is mainly founded on signal processing, machine learning, and deep learning, generalizing over languages continues to remain a challenge. However, developing generalizable and transferable models are critical due to a lack of sufficient resources in terms of data and labels for languages beyond the most commonly spoken ones. To improve performance over languages, we propose a denoising autoencoder with semi-supervision using a continuous metric loss based on either activation or valence. The novelty of this work lies in our proposal of continuous metric learning, which is among the first proposals on the topic to the best of our knowledge. Furthermore, to address the lack of activation and valence labels in the transfer datasets, we annotate the signal samples with activation and valence levels corresponding to a dimensional model of emotions, which were then used to evaluate the quality of the embedding over the transfer datasets. We show that the proposed semi-supervised model consistently outperforms the baseline unsupervised method, which is a conventional denoising autoencoder, in terms of emotion classification accuracy as well as correlation with respect to the dimensional variables. Further evaluation of classification accuracy with respect to the reference, a BERT based speech representation model, shows that the proposed method is comparable to the reference method in classifying specific emotion classes at a much lower complexity.
Complex spectrum and magnitude are considered as two major features of speech enhancement and dereverberation. Traditional approaches always treat these two features separately, ignoring their underlying relationship. In this paper, we proposem Uformer, a Unet based dilated complex & real dual-path conformer network in both complex and magnitude domain for simultaneous speech enhancement and dereverberation. We exploit time attention (TA) and dilated convolution (DC) to leverage local and global contextual information and frequency attention (FA) to model dimensional information. These three sub-modules contained in the proposed dilated complex & real dual-path conformer module effectively improve the speech enhancement and dereverberation performance. Furthermore, hybrid encoder and decoder are adopted to simultaneously model the complex spectrum and magnitude and promote the information interaction between two domains. Encoder decoder attention is also applied to enhance the interaction between encoder and decoder. Our experimental results outperform all SOTA time and complex domain models objectively and subjectively. Specifically, Uformer reaches 3.6032 DNSMOS on the blind test set of Interspeech 2021 DNS Challenge, which outperforms all top-performed models. We also carry out ablation experiments to tease apart all proposed sub-modules that are most important.
At the time of writing, the world population is suffering from more than 10,000 registered COVID-19 disease epidemic induced deaths since the outbreak of the Corona virus more than three months ago now officially known as SARS-CoV-2. Since, tremendous efforts have been made worldwide to counter-steer and control the epidemic by now labelled as pandemic. In this contribution, we provide an overview on the potential for computer audition (CA), i.e., the usage of speech and sound analysis by artificial intelligence to help in this scenario. We first survey which types of related or contextually significant phenomena can be automatically assessed from speech or sound. These include the automatic recognition and monitoring of breathing, dry and wet coughing or sneezing sounds, speech under cold, eating behaviour, sleepiness, or pain to name but a few. Then, we consider potential use-cases for exploitation. These include risk assessment and diagnosis based on symptom histograms and their development over time, as well as monitoring of spread, social distancing and its effects, treatment and recovery, and patient wellbeing. We quickly guide further through challenges that need to be faced for real-life usage. We come to the conclusion that CA appears ready for implementation of (pre-)diagnosis and monitoring tools, and more generally provides rich and significant, yet so far untapped potential in the fight against COVID-19 spread.
Estimating the engagement is critical for human - robot interaction. Engagement measures typically rely on the dynamics of the social signals exchanged by the partners, especially speech and gaze. However, the dynamics of these signals is likely to be influenced by individual and social factors, such as personality traits, as it is well documented that they critically influence how two humans interact with each other. Here, we assess the influence of two factors, namely extroversion and negative attitude toward robots, on speech and gaze during a cooperative task, where a human must physically manipulate a robot to assemble an object. We evaluate if the scores of extroversion and negative attitude towards robots co-variate with the duration and frequency of gaze and speech cues. The experiments were carried out with the humanoid robot iCub and N=56 adult participants. We found that the more people are extrovert, the more and longer they tend to talk with the robot; and the more people have a negative attitude towards robots, the less they will look at the robot face and the more they will look at the robot hands where the assembly and the contacts occur. Our results confirm and provide evidence that the engagement models classically used in human-robot interaction should take into account attitudes and personality traits.
Cross-speaker style transfer (CSST) in text-to-speech (TTS) synthesis aims at transferring a speaking style to the synthesised speech in a target speaker's voice. Most previous CSST approaches rely on expensive high-quality data carrying desired speaking style during training and require a reference utterance to obtain speaking style descriptors as conditioning on the generation of a new sentence. This work presents Referee, a robust reference-free CSST approach for expressive TTS, which fully leverages low-quality data to learn speaking styles from text. Referee is built by cascading a text-to-style (T2S) model with a style-to-wave (S2W) model. Phonetic PosteriorGram (PPG), phoneme-level pitch and energy contours are adopted as fine-grained speaking style descriptors, which are predicted from text using the T2S model. A novel pretrain-refinement method is adopted to learn a robust T2S model by only using readily accessible low-quality data. The S2W model is trained with high-quality target data, which is adopted to effectively aggregate style descriptors and generate high-fidelity speech in the target speaker's voice. Experimental results are presented, showing that Referee outperforms a global-style-token (GST)-based baseline approach in CSST.
Dysarthric speech detection (DSD) systems aim to detect characteristics of the neuromotor disorder from speech. Such systems are particularly susceptible to domain mismatch where the training and testing data come from the source and target domains respectively, but the two domains may differ in terms of speech stimuli, disease etiology, etc. It is hard to acquire labelled data in the target domain, due to high costs of annotating sizeable datasets. This paper makes a first attempt to formulate cross-domain DSD as an unsupervised domain adaptation (UDA) problem. We use labelled source-domain data and unlabelled target-domain data, and propose a multi-task learning strategy, including dysarthria presence classification (DPC), domain adversarial training (DAT) and mutual information minimization (MIM), which aim to learn dysarthria-discriminative and domain-invariant biomarker embeddings. Specifically, DPC helps biomarker embeddings capture critical indicators of dysarthria; DAT forces biomarker embeddings to be indistinguishable in source and target domains; and MIM further reduces the correlation between biomarker embeddings and domain-related cues. By treating the UASPEECH and TORGO corpora respectively as the source and target domains, experiments show that the incorporation of UDA attains absolute increases of 22.2% and 20.0% respectively in utterance-level weighted average recall and speaker-level accuracy.
End-to-end models are favored in automatic speech recognition (ASR) because of its simplified system structure and superior performance. Among these models, recurrent neural network transducer (RNN-T) has achieved significant progress in streaming on-device speech recognition because of its high-accuracy and low-latency. RNN-T adopts a prediction network to enhance language information, but its language modeling ability is limited because it still needs paired speech-text data to train. Further strengthening the language modeling ability through extra text data, such as shallow fusion with an external language model, only brings a small performance gain. In view of the fact that Mandarin Chinese is a character-based language and each character is pronounced as a tonal syllable, this paper proposes a novel cascade RNN-T approach to improve the language modeling ability of RNN-T. Our approach firstly uses an RNN-T to transform acoustic feature into syllable sequence, and then converts the syllable sequence into character sequence through an RNN-T-based syllable-to-character converter. Thus a rich text repository can be easily used to strengthen the language model ability. By introducing several important tricks, the cascade RNN-T approach surpasses the character-based RNN-T by a large margin on several Mandarin test sets, with much higher recognition quality and similar latency.
Voice conversion (VC) aims at conversion of speaker characteristic without altering content. Due to training data limitations and modeling imperfections, it is difficult to achieve believable speaker mimicry without introducing processing artifacts; performance assessment of VC, therefore, usually involves both speaker similarity and quality evaluation by a human panel. As a time-consuming, expensive, and non-reproducible process, it hinders rapid prototyping of new VC technology. We address artifact assessment using an alternative, objective approach leveraging from prior work on spoofing countermeasures (CMs) for automatic speaker verification. Therein, CMs are used for rejecting `fake' inputs such as replayed, synthetic or converted speech but their potential for automatic speech artifact assessment remains unknown. This study serves to fill that gap. As a supplement to subjective results for the 2018 Voice Conversion Challenge (VCC'18) data, we configure a standard constant-Q cepstral coefficient CM to quantify the extent of processing artifacts. Equal error rate (EER) of the CM, a confusability index of VC samples with real human speech, serves as our artifact measure. Two clusters of VCC'18 entries are identified: low-quality ones with detectable artifacts (low EERs), and higher quality ones with less artifacts. None of the VCC'18 systems, however, is perfect: all EERs are < 30 % (the `ideal' value would be 50 %). Our preliminary findings suggest potential of CMs outside of their original application, as a supplemental optimization and benchmarking tool to enhance VC technology.
Social media has become a bedrock for people to voice their opinions worldwide. Due to the greater sense of freedom with the anonymity feature, it is possible to disregard social etiquette online and attack others without facing severe consequences, inevitably propagating hate speech. The current measures to sift the online content and offset the hatred spread do not go far enough. One factor contributing to this is the prevalence of regional languages in social media and the paucity of language flexible hate speech detectors. The proposed work focuses on analyzing hate speech in Hindi-English code-switched language. Our method explores transformation techniques to capture precise text representation. To contain the structure of data and yet use it with existing algorithms, we developed MoH or Map Only Hindi, which means "Love" in Hindi. MoH pipeline consists of language identification, Roman to Devanagari Hindi transliteration using a knowledge base of Roman Hindi words. Finally, it employs the fine-tuned Multilingual Bert and MuRIL language models. We conducted several quantitative experiment studies on three datasets and evaluated performance using Precision, Recall, and F1 metrics. The first experiment studies MoH mapped text's performance with classical machine learning models and shows an average increase of 13% in F1 scores. The second compares the proposed work's scores with those of the baseline models and offers a rise in performance by 6%. Finally, the third reaches the proposed MoH technique with various data simulations using the existing transliteration library. Here, MoH outperforms the rest by 15%. Our results demonstrate a significant improvement in the state-of-the-art scores on all three datasets.