Phase retrieval in its most general form is the problem of reconstructing a complex valued function from phaseless information of some transform of that function. This problem arises in various fields such as X-ray crystallography, electron microscopy, coherent diffractive imaging, astronomy, speech recognition, and quantum mechanics. The mathematical and computational analysis of these problems has a long history and a variety of different algorithms has been proposed in the literature. The performance of which usually depends on the constraints imposed on the sought function and the number of measurements. In this paper, we present an algorithm for coherent diffractive imaging with phaseless measurements. The algorithm accounts for both coherent and incoherent wave propagation and allows for reconstructing absorption as well as phase images that quantify the attenuation and the refraction of the waves when they go through an object. The algorithm requires coherent or partially coherent illumination, and several detectors to record the intensity of the distorted wave that passes through the object under inspection. To obtain enough information for imaging, a series of masks are introduced between the source and the object that create a diversity of illumination patterns.
To accomplish the punctuation restoration task, most existing approaches focused on leveraging extra information (e.g., part-of-speech tags) or addressing the class imbalance problem. Recent works have widely applied the transformer-based language models and significantly improved their effectiveness. To the best of our knowledge, an inherent issue has remained neglected: the attention of individual heads in the transformer will be diluted or powerless while feeding the long non-punctuation utterances. Since those previous contexts, not the followings, are comparatively more valuable to the current position, it's hard to achieve a good balance by independent attention. In this paper, we propose a novel Feature Fusion framework based on two-type Attentions (FFA) to alleviate the shortage. It introduces a two-stream architecture. One module involves interaction between attention heads to encourage the communication, and another masked attention module captures the dependent feature representation. Then, it aggregates two feature embeddings to fuse information and enhances context-awareness. The experiments on the popular benchmark dataset IWSLT demonstrate that our approach is effective. Without additional data, it obtains comparable performance to the current state-of-the-art models.
Recent research on speech enhancement (SE) has seen the emergence of deep learning-based methods. It is still a challenging task to determine effective ways to increase the generalizability of SE under diverse test conditions. In this paper, we combine zero-shot learning and ensemble learning to propose a zero-shot model selection (ZMOS) approach to increase the generalization of SE performance. The proposed approach is realized in two phases, namely offline and online phases. The offline phase clusters the entire set of training data into multiple subsets, and trains a specialized SE model (termed component SE model) with each subset. The online phase selects the most suitable component SE model to carry out enhancement. Two selection strategies are developed: selection based on quality score (QS) and selection based on quality embedding (QE). Both QS and QE are obtained by a Quality-Net, a non-intrusive quality assessment network. In the offline phase, the QS or QE of a train-ing utterance is used to group the training data into clusters. In the online phase, the QS or QE of the test utterance is used to identify the appropriate component SE model to perform enhancement on the test utterance. Experimental results have confirmed that the proposed ZMOS approach can achieve better performance in both seen and unseen noise types compared to the baseline systems, which indicates the effectiveness of the proposed approach to provide robust SE performance.
Error correction techniques have been used to refine the output sentences from automatic speech recognition (ASR) models and achieve a lower word error rate (WER) than original ASR outputs. Previous works usually use a sequence-to-sequence model to correct an ASR output sentence autoregressively, which causes large latency and cannot be deployed in online ASR services. A straightforward solution to reduce latency, inspired by non-autoregressive (NAR) neural machine translation, is to use an NAR sequence generation model for ASR error correction, which, however, comes at the cost of significantly increased ASR error rate. In this paper, observing distinctive error patterns and correction operations (i.e., insertion, deletion, and substitution) in ASR, we propose FastCorrect, a novel NAR error correction model based on edit alignment. In training, FastCorrect aligns each source token from an ASR output sentence to the target tokens from the corresponding ground-truth sentence based on the edit distance between the source and target sentences, and extracts the number of target tokens corresponding to each source token during edition/correction, which is then used to train a length predictor and to adjust the source tokens to match the length of the target sentence for parallel generation. In inference, the token number predicted by the length predictor is used to adjust the source tokens for target sequence generation. Experiments on the public AISHELL-1 dataset and an internal industrial-scale ASR dataset show the effectiveness of FastCorrect for ASR error correction: 1) it speeds up the inference by 6-9 times and maintains the accuracy (8-14% WER reduction) compared with the autoregressive correction model; and 2) it outperforms the accuracy of popular NAR models adopted in neural machine translation by a large margin.
Multi-source localization is an important and challenging technique for multi-talker conversation analysis. This paper proposes a novel supervised learning method using deep neural networks to estimate the direction of arrival (DOA) of all the speakers simultaneously from the audio mixture. At the heart of the proposal is a source splitting mechanism that creates source-specific intermediate representations inside the network. This allows our model to give source-specific posteriors as the output unlike the traditional multi-label classification approach. Existing deep learning methods perform a frame level prediction, whereas our approach performs an utterance level prediction by incorporating temporal selection and averaging inside the network to avoid post-processing. We also experiment with various loss functions and show that a variant of earth mover distance (EMD) is very effective in classifying DOA at a very high resolution by modeling inter-class relationships. In addition to using the prediction error as a metric for evaluating our localization model, we also establish its potency as a frontend with automatic speech recognition (ASR) as the downstream task. We convert the estimated DOAs into a feature suitable for ASR and pass it as an additional input feature to a strong multi-channel and multi-talker speech recognition baseline. This added input feature drastically improves the ASR performance and gives a word error rate (WER) of 6.3% on the evaluation data of our simulated noisy two speaker mixtures, while the baseline which doesn't use explicit localization input has a WER of 11.5%. We also perform ASR evaluation on real recordings with the overlapped set of the MC-WSJ-AV corpus in addition to simulated mixtures.
Neural evaluation metrics derived for numerous speech generation tasks have recently attracted great attention. In this paper, we propose SVSNet, the first end-to-end neural network model to assess the speaker voice similarity between natural speech and synthesized speech. Unlike most neural evaluation metrics that use hand-crafted features, SVSNet directly takes the raw waveform as input to more completely utilize speech information for prediction. SVSNet consists of encoder, co-attention, distance calculation, and prediction modules and is trained in an end-to-end manner. The experimental results on the Voice Conversion Challenge 2018 and 2020 (VCC2018 and VCC2020) datasets show that SVSNet notably outperforms well-known baseline systems in the assessment of speaker similarity at the utterance and system levels.
Animal vocalisations and natural soundscapes are fascinating objects of study, and contain valuable evidence about animal behaviours, populations and ecosystems. They are studied in bioacoustics and ecoacoustics, with signal processing and analysis an important component. Computational bioacoustics has accelerated in recent decades due to the growth of affordable digital sound recording devices, and to huge progress in informatics such as big data, signal processing and machine learning. Methods are inherited from the wider field of deep learning, including speech and image processing. However, the tasks, demands and data characteristics are often different from those addressed in speech or music analysis. There remain unsolved problems, and tasks for which evidence is surely present in many acoustic signals, but not yet realised. In this paper I perform a review of the state of the art in deep learning for computational bioacoustics, aiming to clarify key concepts and identify and analyse knowledge gaps. Based on this, I offer a subjective but principled roadmap for computational bioacoustics with deep learning: topics that the community should aim to address, in order to make the most of future developments in AI and informatics, and to use audio data in answering zoological and ecological questions.
Speech rhythms have been dealt with in three main ways: from the introspective analyses of rhythm as a correlate of syllable and foot timing in linguistics and applied linguistics, through analyses of durations of segments of utterances associated with consonantal and vocalic properties, syllables, feet and words, to models of rhythms in speech production and perception as physical oscillations. The present study avoids introspection and human-filtered annotation methods and extends the signal processing paradigm of amplitude envelope spectrum analysis by adding an additional analytic step of edge detection, and postulating the co-existence of multiple speech rhythms in rhythm zones marked by identifiable edges (Rhythm Zone Theory, RZT). An exploratory investigation of the utility of RZT is conducted, suggesting that native and non-native readings of the same text are distinct sub-genres of read speech: a reading by a US native speaker and non-native readings by relatively low-performing Cantonese adult learners of English. The study concludes by noting that with the methods used, RZT can distinguish between the speech rhythms of well-defined sub-genres of native speaker reading vs. non-native learner reading, but needs further refinement in order to be applied to the paradoxically more complex speech of low-performing language learners, whose speech rhythms are co-determined by non-fluency and disfluency factors in addition to well-known linguistic factors of grammar, vocabulary and discourse constraints.
Personal narratives (PN) - spoken or written - are recollections of facts, people, events, and thoughts from one's own experience. Emotion recognition and sentiment analysis tasks are usually defined at the utterance or document level. However, in this work, we focus on Emotion Carriers (EC) defined as the segments (speech or text) that best explain the emotional state of the narrator ("loss of father", "made me choose"). Once extracted, such EC can provide a richer representation of the user state to improve natural language understanding and dialogue modeling. In previous work, it has been shown that EC can be identified using lexical features. However, spoken narratives should provide a richer description of the context and the users' emotional state. In this paper, we leverage word-based acoustic and textual embeddings as well as early and late fusion techniques for the detection of ECs in spoken narratives. For the acoustic word-level representations, we use Residual Neural Networks (ResNet) pretrained on separate speech emotion corpora and fine-tuned to detect EC. Experiments with different fusion and system combination strategies show that late fusion leads to significant improvements for this task.
Current text-to-speech algorithms produce realistic fakes of human voices, making deepfake detection a much-needed area of research. While researchers have presented various techniques for detecting audio spoofs, it is often unclear exactly why these architectures are successful: Preprocessing steps, hyperparameter settings, and the degree of fine-tuning are not consistent across related work. Which factors contribute to success, and which are accidental? In this work, we address this problem: We systematize audio spoofing detection by re-implementing and uniformly evaluating architectures from related work. We identify overarching features for successful audio deepfake detection, such as using cqtspec or logspec features instead of melspec features, which improves performance by 37% EER on average, all other factors constant. Additionally, we evaluate generalization capabilities: We collect and publish a new dataset consisting of 37.9 hours of found audio recordings of celebrities and politicians, of which 17.2 hours are deepfakes. We find that related work performs poorly on such real-world data (performance degradation of up to one thousand percent). This may suggest that the community has tailored its solutions too closely to the prevailing ASVSpoof benchmark and that deepfakes are much harder to detect outside the lab than previously thought.