Neural models have become ubiquitous in automatic speech recognition systems. While neural networks are typically used as acoustic models in more complex systems, recent studies have explored end-to-end speech recognition systems based on neural networks, which can be trained to directly predict text from input acoustic features. Although such systems are conceptually elegant and simpler than traditional systems, it is less obvious how to interpret the trained models. In this work, we analyze the speech representations learned by a deep end-to-end model that is based on convolutional and recurrent layers, and trained with a connectionist temporal classification (CTC) loss. We use a pre-trained model to generate frame-level features which are given to a classifier that is trained on frame classification into phones. We evaluate representations from different layers of the deep model and compare their quality for predicting phone labels. Our experiments shed light on important aspects of the end-to-end model such as layer depth, model complexity, and other design choices.
State of the art time automatic speech recognition (ASR) systems are becoming increasingly complex and expensive for practical applications. This paper presents the development of a high performance and low-footprint 4-bit quantized LF-MMI trained factored time delay neural networks (TDNNs) based ASR system on the 300-hr Switchboard corpus. A key feature of the overall system design is to account for the fine-grained, varying performance sensitivity at different model components to quantization errors. To this end, a set of neural architectural compression and mixed precision quantization approaches were used to facilitate hidden layer level auto-configuration of optimal factored TDNN weight matrix subspace dimensionality and quantization bit-widths. The proposed techniques were also used to produce 2-bit mixed precision quantized Transformer language models. Experiments conducted on the Switchboard data suggest that the proposed neural architectural compression and mixed precision quantization techniques consistently outperform the uniform precision quantised baseline systems of comparable bit-widths in terms of word error rate (WER). An overall "lossless" compression ratio of 13.6 was obtained over the baseline full precision system including both the TDNN and Transformer components while incurring no statistically significant WER increase.
Fundamental modelling differences between hybrid and end-to-end (E2E) automatic speech recognition (ASR) systems create large diversity and complementarity among them. This paper investigates multi-pass rescoring and cross adaptation based system combination approaches for hybrid TDNN and Conformer E2E ASR systems. In multi-pass rescoring, state-of-the-art hybrid LF-MMI trained CNN-TDNN system featuring speed perturbation, SpecAugment and Bayesian learning hidden unit contributions (LHUC) speaker adaptation was used to produce initial N-best outputs before being rescored by the speaker adapted Conformer system using a 2-way cross system score interpolation. In cross adaptation, the hybrid CNN-TDNN system was adapted to the 1-best output of the Conformer system or vice versa. Experiments on the 300-hour Switchboard corpus suggest that the combined systems derived using either of the two system combination approaches outperformed the individual systems. The best combined system obtained using multi-pass rescoring produced statistically significant word error rate (WER) reductions of 2.5% to 3.9% absolute (22.5% to 28.9% relative) over the stand alone Conformer system on the NIST Hub5'00, Rt03 and Rt02 evaluation data.
We study the effectiveness of several techniques to personalize end-to-end speech models and improve the recognition of proper names relevant to the user. These techniques differ in the amounts of user effort required to provide supervision, and are evaluated on how they impact speech recognition performance. We propose using keyword-dependent precision and recall metrics to measure vocabulary acquisition performance. We evaluate the algorithms on a dataset that we designed to contain names of persons that are difficult to recognize. Therefore, the baseline recall rate for proper names in this dataset is very low: 2.4%. A data synthesis approach we developed brings it to 48.6%, with no need for speech input from the user. With speech input, if the user corrects only the names, the name recall rate improves to 64.4%. If the user corrects all the recognition errors, we achieve the best recall of 73.5%. To eliminate the need to upload user data and store personalized models on a server, we focus on performing the entire personalization workflow on a mobile device.
Recently, language identity information has been utilized to improve the performance of end-to-end code-switching (CS) speech recognition. However, previous works use an additional language identification (LID) model as an auxiliary module, which causes the system complex. In this work, we propose an improved recurrent neural network transducer (RNN-T) model with language bias to alleviate the problem. We use the language identities to bias the model to predict the CS points. This promotes the model to learn the language identity information directly from transcription, and no additional LID model is needed. We evaluate the approach on a Mandarin-English CS corpus SEAME. Compared to our RNN-T baseline, the proposed method can achieve 16.2% and 12.9% relative error reduction on two test sets, respectively.
We propose an algorithm to explore the global optimization method, using SAT solvers, for training a neural net. Deep Neural Networks have achieved great feats in tasks like-image recognition, speech recognition, etc. Much of their success can be attributed to the gradient-based optimisation methods, which scale well to huge datasets while still giving solutions, better than any other existing methods. However, there exist learning problems like the parity function and the Fast Fourier Transform, where a neural network using gradient-based optimisation algorithm can not capture the underlying structure of the learning task properly. Thus, exploring global optimisation methods is of utmost interest as the gradient-based methods get stuck in local optima. In the experiments, we demonstrate the effectiveness of our algorithm against the ADAM optimiser in certain tasks like parity learning. However, in the case of image classification on the MNIST Dataset, the performance of our algorithm was less than satisfactory. We further discuss the role of the size of the training dataset and the hyper-parameter settings in keeping things scalable for a SAT solver.
Speech enhancement has seen great improvement in recent years using end-to-end neural networks. However, most models are agnostic to the spoken phonetic content. Recently, several studies suggested phonetic-aware speech enhancement, mostly using perceptual supervision. Yet, injecting phonetic features during model optimization can take additional forms (e.g., model conditioning). In this paper, we conduct a systematic comparison between different methods of incorporating phonetic information in a speech enhancement model. By conducting a series of controlled experiments, we observe the influence of different phonetic content models as well as various feature-injection techniques on enhancement performance, considering both causal and non-causal models. Specifically, we evaluate three settings for injecting phonetic information, namely: i) feature conditioning; ii) perceptual supervision; and iii) regularization. Phonetic features are obtained using an intermediate layer of either a supervised pre-trained Automatic Speech Recognition (ASR) model or by using a pre-trained Self-Supervised Learning (SSL) model. We further observe the effect of choosing different embedding layers on performance, considering both manual and learned configurations. Results suggest that using a SSL model as phonetic features outperforms the ASR one in most cases. Interestingly, the conditioning setting performs best among the evaluated configurations.
Speech applications dealing with conversations require not only recognizing the spoken words, but also determining who spoke when. The task of assigning words to speakers is typically addressed by merging the outputs of two separate systems, namely, an automatic speech recognition (ASR) system and a speaker diarization (SD) system. The two systems are trained independently with different objective functions. Often the SD systems operate directly on the acoustics and are not constrained to respect word boundaries and this deficiency is overcome in an ad hoc manner. Motivated by recent advances in sequence to sequence learning, we propose a novel approach to tackle the two tasks by a joint ASR and SD system using a recurrent neural network transducer. Our approach utilizes both linguistic and acoustic cues to infer speaker roles, as opposed to typical SD systems, which only use acoustic cues. We evaluated the performance of our approach on a large corpus of medical conversations between physicians and patients. Compared to a competitive conventional baseline, our approach improves word-level diarization error rate from 15.8% to 2.2%.
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.
The majority of existing speech emotion recognition models are trained and evaluated on a single corpus and a single language setting. These systems do not perform as well when applied in a cross-corpus and cross-language scenario. This paper presents results for speech emotion recognition for 4 languages in both single corpus and cross corpus setting. Additionally, since multi-task learning (MTL) with gender, naturalness and arousal as auxiliary tasks has shown to enhance the generalisation capabilities of the emotion models, this paper introduces language ID as another auxiliary task in MTL framework to explore the role of spoken language on emotion recognition which has not been studied yet.