We test the robustness of a maximum-likelihood (ML) based classifier where sequential data as observation is corrupted by noise. The hypothesis is that a generative model, that combines the state transitions of a hidden Markov model (HMM) and the neural network based probability distributions for the hidden states of the HMM, can provide a robust classification performance. The combined model is called normalizing-flow mixture model based HMM (NMM-HMM). It can be trained using a combination of expectation-maximization (EM) and backpropagation. We verify the improved robustness of NMM-HMM classifiers in an application to speech recognition.
Effectiveness of teaching digital signal processing can be enhanced by reducing lecture time devoted to theory, and increasing emphasis on applications, programming aspects, visualization and intuitive understanding. An integrated approach to teaching requires instructors to simultaneously teach theory and its applications in storage and processing of audio, speech and biomedical signals. Student engagement can be enhanced by engaging students to work in groups during the class where students can solve short problems and short programming assignments or take quizzes. These approaches will increase student interest in learning the subject and student engagement.
Automatic Speech Recognition (ASR) systems introduce word errors, which often confuse punctuation prediction models, turning punctuation restoration into a challenging task. These errors usually take the form of homonyms. We show how retrofitting of the word embeddings on the domain-specific data can mitigate ASR errors. Our main contribution is a method for better alignment of homonym embeddings and the validation of the presented method on the punctuation prediction task. We record the absolute improvement in punctuation prediction accuracy between 6.2% (for question marks) to 9% (for periods) when compared with the state-of-the-art model.
Recent character and phoneme-based parametric TTS systems using deep learning have shown strong performance in natural speech generation. However, the choice between character or phoneme input can create serious limitations for practical deployment, as direct control of pronunciation is crucial in certain cases. We demonstrate a simple method for combining multiple types of linguistic information in a single encoder, named representation mixing, enabling flexible choice between character, phoneme, or mixed representations during inference. Experiments and user studies on a public audiobook corpus show the efficacy of our approach.
While part-of-speech (POS) tagging and dependency parsing are observed to be closely related, existing work on joint modeling with manually crafted feature templates suffers from the feature sparsity and incompleteness problems. In this paper, we propose an approach to joint POS tagging and dependency parsing using transition-based neural networks. Three neural network based classifiers are designed to resolve shift/reduce, tagging, and labeling conflicts. Experiments show that our approach significantly outperforms previous methods for joint POS tagging and dependency parsing across a variety of natural languages.
Separation of competing speech is a key challenge in signal processing and a feat routinely performed by the human auditory brain. A long standing benchmark of the spectrogram approach to source separation is known as the ideal binary mask. Here, we train a convolutional deep neural network, on a two-speaker cocktail party problem, to make probabilistic predictions about binary masks. Our results approach ideal binary mask performance, illustrating that relatively simple deep neural networks are capable of robust binary mask prediction. We also illustrate the trade-off between prediction statistics and separation quality.
The objective of change-point detection is to discover abrupt property changes lying behind time-series data. In this paper, we present a novel statistical change-point detection algorithm based on non-parametric divergence estimation between time-series samples from two retrospective segments. Our method uses the relative Pearson divergence as a divergence measure, and it is accurately and efficiently estimated by a method of direct density-ratio estimation. Through experiments on artificial and real-world datasets including human-activity sensing, speech, and Twitter messages, we demonstrate the usefulness of the proposed method.
A new three-stage computer artificial neural network model of the tip-of-the-tongue phenomenon is shortly described, and its stochastic nature was demonstrated. A way to calculate strength and appearance probability of tip-of-the-tongue states, neural network mechanism of feeling-of-knowing phenomenon are proposed. The model synthesizes memory, psycholinguistic, and metamemory approaches, bridges speech errors and naming chronometry research traditions. A model analysis of a tip-of-the-tongue case from Anton Chekhov's short story 'A Horsey Name' is performed. A new 'throw-up-one's-arms effect' is defined.
In recent years, multimodal AI has seen an upward trend as researchers are integrating data of different types such as text, images, speech into modelling to get the best results. This project leverages multimodal AI and matrix factorization techniques for representation learning, on text and image data simultaneously, thereby employing the widely used techniques of Natural Language Processing (NLP) and Computer Vision. The learnt representations are evaluated using downstream classification and regression tasks. The methodology adopted can be extended beyond the scope of this project as it uses Auto-Encoders for unsupervised representation learning.
In this paper we propose the inversion of nonlinear distortions in order to improve the recognition rates of a speaker recognizer system. We study the effect of saturations on the test signals, trying to take into account real situations where the training material has been recorded in a controlled situation but the testing signals present some mismatch with the input signal level (saturations). The experimental results shows that a combination of data fusion with and without nonlinear distortion compensation can improve the recognition rates with saturated test sentences from 80% to 88.57%, while the results with clean speech (without saturation) is 87.76% for one microphone.