Second language (L2) speech is often labeled with the native, phone categories. However, in many cases, it is difficult to decide on a categorical phone that an L2 segment belongs to. These segments are regarded as non-categories. Most existing approaches for Mispronunciation Detection and Diagnosis (MDD) are only concerned with categorical errors, i.e. a phone category is inserted, deleted or substituted by another. However, non-categorical errors are not considered. To model these non-categorical errors, this work aims at exploring non-categorical patterns to extend the categorical phone set. We apply a phonetic segment classifier to generate segmental phonetic posterior-grams (SPPGs) to represent phone segment-level information. And then we explore the non-categories by looking for the SPPGs with more than one peak. Compared with the baseline system, this approach explores more non-categorical patterns, and also perceptual experimental results show that the explored non-categories are more accurate with increased confusion degree by 7.3% and 7.5% under two different measures. Finally, we preliminarily analyze the reason behind those non-categories.
Toyota Technological Institute at Chicago（芝加哥丰田技术研究所）
Discriminative segmental models offer a way to incorporate flexible feature functions into speech recognition. However, their appeal has been limited by their computational requirements, due to the large number of possible segments to consider. Multi-pass cascades of segmental models introduce features of increasing complexity in different passes, where in each pass a segmental model rescores lattices produced by a previous (simpler) segmental model. In this paper, we explore several ways of making segmental cascades efficient and practical: reducing the feature set in the first pass, frame su…
Toyota Technological Institute at Chicago（芝加哥丰田技术研究所）
Discriminative segmental models, such as segmental conditional random fields (SCRFs) and segmental structured support vector machines (SSVMs), have had success in speech recognition via both lattice rescoring and first-pass decoding. However, such models suffer from slow decoding, hampering the use of computationally expensive features, such as segment neural networks or other high-order features. A typical solution is to use approximate decoding, either by beam pruning in a single pass or by beam pruning to generate a lattice followed by a second pass. In this work, we study discriminative se…
While discrete latent variable models have had great success in self-supervised learning, most models assume that frames are independent. Due to the segmental nature of phonemes in speech perception, modeling dependencies among latent variables at the frame level can potentially improve the learned representations on phonetic-related tasks. In this work, we assume Markovian dependencies among latent variables, and propose to learn speech representations with neural hidden Markov models. Our general framework allows us to compare to self-supervised models that assume independence, while keeping…
Idiap Research Institute（Idiap研究所）；École Polytechnique Fédérale de Lausanne (EPFL)（洛桑联邦理工学院）
The speech signal conveys information on different time scales from short time scale or segmental, associated to phonological and phonetic information to long time scale or supra segmental, associated to syllabic and prosodic information. Linguistic and neurocognitive studies recognize the phonological classes at segmental level as the essential and invariant representations used in speech temporal organization. In the context of speech processing, a deep neural network (DNN) is an effective computational method to infer the probability of individual phonological classes from a short segment o…
University of Groningen（格罗宁根大学）；University of Pennsylvania（宾夕法尼亚大学）；Haskins Laboratories（哈斯金斯实验室）
Variation in speech is often quantified by comparing phonetic transcriptions of the same utterance. However, manually transcribing speech is time-consuming and error prone. As an alternative, therefore, we investigate the extraction of acoustic embeddings from several self-supervised neural models. We use these representations to compute word-based pronunciation differences between non-native and native speakers of English, and between Norwegian dialect speakers. For comparison with several earlier studies, we evaluate how well these differences match human perception by comparing them with av…
The present study has two goals relating to the grammar of prosody, understood as the rhythms and melodies of speech. First, an overview is provided of the computable grammatical and phonetic approaches to prosody analysis which use hypothetico-deductive methods and are based on learned hermeneutic intuitions about language. Second, a proposal is presented for an inductive grounding in the physical signal, in which prosodic structure is inferred using a language-independent method from the low-frequency spectrum of the speech signal. The overview includes a discussion of computational aspects …
Most current very low bit rate (VLBR) speech coding systems use hidden Markov model (HMM) based speech recognition/synthesis techniques. This allows transmission of information (such as phonemes) segment by segment that decreases the bit rate. However, the encoder based on a phoneme speech recognition may create bursts of segmental errors. Segmental errors are further propagated to optional suprasegmental (such as syllable) information coding. Together with the errors of voicing detection in pitch parametrization, HMM-based speech coding creates speech discontinuities and unnatural speech soun…
Google, Inc.（谷歌公司）；University of Washington（华盛顿大学）
The paper revives an older approach to acoustic modeling that borrows from n-gram language modeling in an attempt to scale up both the amount of training data and model size (as measured by the number of parameters in the model), to approximately 100 times larger than current sizes used in automatic speech recognition. In such a data-rich setting, we can expand the phonetic context significantly beyond triphones, as well as increase the number of Gaussian mixture components for the context-dependent states that allow it. We have experimented with contexts that span seven or more context-indepe…
Speech processing systems rely on robust feature extraction to handle phonetic and semantic variations found in natural language. While techniques exist for desensitizing features to common noise patterns produced by Speech-to-Text (STT) and Text-to-Speech (TTS) systems, the question remains how to best leverage state-of-the-art language models (which capture rich semantic features, but are trained on only written text) on inputs with ASR errors. In this paper, we present Telephonetic, a data augmentation framework that helps robustify language model features to ASR corrupted inputs. To captur…
Islamic Azad university Science and research branch-khuzestan（伊斯兰阿扎德大学科学与研究分支（胡齐斯坦））；Amirkabir University of Technology（阿米尔卡比尔理工大学）
The interest in text to speech synthesis increased in the world .text to speech have been developed formany popular languages such as English, Spanish and French and many researches and developmentshave been applied to those languages. Persian on the other hand, has been given little attentioncompared to other languages of similar importance and the research in Persian is still in its infancy.Persian language possess many difficulty and exceptions that increase complexity of text to speechsystems. For example: short vowels is absent in written text or existence of homograph words. in thispaper…
University of Edinburgh（爱丁堡大学）；Toyota Technological Institute at Chicago（芝加哥丰田技术研究所）
Recent studies have been revisiting whole words as the basic modelling unit in speech recognition and query applications, instead of phonetic units. Such whole-word segmental systems rely on a function that maps a variable-length speech segment to a vector in a fixed-dimensional space; the resulting acoustic word embeddings need to allow for accurate discrimination between different word types, directly in the embedding space. We compare several old and new approaches in a word discrimination task. Our best approach uses side information in the form of known word pairs to train a Siamese convo…
Advancing Sentiment Analysis in Tamil-English Code-Mixed Texts: Challenges and Transformer-Based Solutions
Instituto Politécnico Nacional (IPN)（墨西哥国立理工学院）；Centro de Investigación en Computación (CIC)（计算机研究中心）；Escuela Superior de Ingeniería Mecánica y Eléctrica (ESIME)（机械与电气工程高等学院）
Automatic speech recognition (ASR) has been widely researched with supervised approaches, while many low-resourced languages lack audio-text aligned data, and supervised methods cannot be applied on them. In this work, we propose a framework to achieve unsupervised ASR on a read English speech dataset, where audio and text are unaligned. In the first stage, each word-level audio segment in the utterances is represented by a vector representation extracted by a sequence-of-sequence autoencoder, in which phonetic information and speaker information are disentangled. Secondly, semantic embeddings…
Post-editing in Automatic Speech Recognition (ASR) entails automatically correcting common and systematic errors produced by the ASR system. The outputs of an ASR system are largely prone to phonetic and spelling errors. In this paper, we propose to use a powerful pre-trained sequence-to-sequence model, BART, further adaptively trained to serve as a denoising model, to correct errors of such types. The adaptive training is performed on an augmented dataset obtained by synthetically inducing errors as well as by incorporating actual errors from an existing ASR system. We also propose a simple a…
