We have recently shown that deep Long Short-Term Memory (LSTM) recurrent neural networks (RNNs) outperform feed forward deep neural networks (DNNs) as acoustic models for speech recognition. More recently, we have shown that the performance of sequence trained context dependent (CD) hidden Markov model (HMM) acoustic models using such LSTM RNNs can be equaled by sequence trained phone models initialized with connectionist temporal classification (CTC). In this paper, we present techniques that further improve performance of LSTM RNN acoustic models for large vocabulary speech recognition. We show that frame stacking and reduced frame rate lead to more accurate models and faster decoding. CD phone modeling leads to further improvements. We also present initial results for LSTM RNN models outputting words directly.
Despite the significant progress made in the last years, state-of-the-art speech recognition technologies provide a satisfactory performance only in the close-talking condition. Robustness of distant speech recognition in adverse acoustic conditions, on the other hand, remains a crucial open issue for future applications of human-machine interaction. To this end, several advances in speech enhancement, acoustic scene analysis as well as acoustic modeling, have recently contributed to improve the state-of-the-art in the field. One of the most effective approaches to derive a robust acoustic modeling is based on using contaminated speech, which proved helpful in reducing the acoustic mismatch between training and testing conditions. In this paper, we revise this classical approach in the context of modern DNN-HMM systems, and propose the adoption of three methods, namely, asymmetric context windowing, close-talk based supervision, and close-talk based pre-training. The experimental results, obtained using both real and simulated data, show a significant advantage in using these three methods, overall providing a 15% error rate reduction compared to the baseline systems. The same trend in performance is confirmed either using a high-quality training set of small size, and a large one.
Online speech recognition is crucial for developing natural human-machine interfaces. This modality, however, is significantly more challenging than off-line ASR, since real-time/low-latency constraints inevitably hinder the use of future information, that is known to be very helpful to perform robust predictions. A popular solution to mitigate this issue consists of feeding neural acoustic models with context windows that gather some future frames. This introduces a latency which depends on the number of employed look-ahead features. This paper explores a different approach, based on estimating the future rather than waiting for it. Our technique encourages the hidden representations of a unidirectional recurrent network to embed some useful information about the future. Inspired by a recently proposed technique called Twin Networks, we add a regularization term that forces forward hidden states to be as close as possible to cotemporal backward ones, computed by a "twin" neural network running backwards in time. The experiments, conducted on a number of datasets, recurrent architectures, input features, and acoustic conditions, have shown the effectiveness of this approach. One important advantage is that our method does not introduce any additional computation at test time if compared to standard unidirectional recurrent networks.
Very deep CNNs achieve state-of-the-art results in both computer vision and speech recognition, but are difficult to train. The most popular way to train very deep CNNs is to use shortcut connections (SC) together with batch normalization (BN). Inspired by Self-Normalizing Neural Networks, we propose the self-normalizing deep CNN (SNDCNN) based acoustic model topology, by removing the SC/BN and replacing the typical RELU activations with scaled exponential linear unit (SELU) in ResNet-50. SELU activations make the network self-normalizing and remove the need for both shortcut connections and batch normalization. Compared to ResNet-50, we can achieve the same or lower word error rate (WER) while at the same time improving both training and inference speed by 60%-80%. We also explore other model inference optimizations to further reduce latency for production use.
With 4.5 million hours of English speech from 10 different sources across 120 countries and models of up to 10 billion parameters, we explore the frontiers of scale for automatic speech recognition. We propose data selection techniques to efficiently scale training data to find the most valuable samples in massive datasets. To efficiently scale model sizes, we leverage various optimizations such as sparse transducer loss and model sharding. By training 1-10B parameter universal English ASR models, we push the limits of speech recognition performance across many domains. Furthermore, our models learn powerful speech representations with zero and few-shot capabilities on novel domains and styles of speech, exceeding previous results across multiple in-house and public benchmarks. For speakers with disorders due to brain damage, our best zero-shot and few-shot models achieve 22% and 60% relative improvement on the AphasiaBank test set, respectively, while realizing the best performance on public social media videos. Furthermore, the same universal model reaches equivalent performance with 500x less in-domain data on the SPGISpeech financial-domain dataset.
Recently, Conformer based CTC/AED model has become a mainstream architecture for ASR. In this paper, based on our prior work, we identify and integrate several approaches to achieve further improvements for ASR tasks, which we denote as multi-loss, multi-path and multi-level, summarized as "3M" model. Specifically, multi-loss refers to the joint CTC/AED loss and multi-path denotes the Mixture-of-Experts(MoE) architecture which can effectively increase the model capacity without remarkably increasing computation cost. Multi-level means that we introduce auxiliary loss at multiple level of a deep model to help training. We evaluate our proposed method on the public WenetSpeech dataset and experimental results show that the proposed method provides 12.2%-17.6% relative CER improvement over the baseline model trained by Wenet toolkit. On our large scale dataset of 150k hours corpus, the 3M model has also shown obvious superiority over the baseline Conformer model.
Several end-to-end deep learning approaches have been recently presented which extract either audio or visual features from the input images or audio signals and perform speech recognition. However, research on end-to-end audiovisual models is very limited. In this work, we present an end-to-end audiovisual model based on residual networks and Bidirectional Gated Recurrent Units (BGRUs). To the best of our knowledge, this is the first audiovisual fusion model which simultaneously learns to extract features directly from the image pixels and audio waveforms and performs within-context word recognition on a large publicly available dataset (LRW). The model consists of two streams, one for each modality, which extract features directly from mouth regions and raw waveforms. The temporal dynamics in each stream/modality are modeled by a 2-layer BGRU and the fusion of multiple streams/modalities takes place via another 2-layer BGRU. A slight improvement in the classification rate over an end-to-end audio-only and MFCC-based model is reported in clean audio conditions and low levels of noise. In presence of high levels of noise, the end-to-end audiovisual model significantly outperforms both audio-only models.
Driven by deep learning techniques and large-scale datasets, recent years have witnessed a paradigm shift in automatic lip reading. While the main thrust of Visual Speech Recognition (VSR) was improving accuracy of Audio Speech Recognition systems, other potential applications, such as biometric identification, and the promised gains of VSR systems, have motivated extensive efforts on developing the lip reading technology. This paper provides a comprehensive survey of the state-of-the-art deep learning based VSR research with a focus on data challenges, task-specific complications, and the corresponding solutions. Advancements in these directions will expedite the transformation of silent speech interface from theory to practice. We also discuss the main modules of a VSR pipeline and the influential datasets. Finally, we introduce some typical VSR application concerns and impediments to real-world scenarios as well as future research directions.
End-to-end automatic speech recognition (ASR) directly maps input speech to a character sequence without using pronunciation lexica. However, in languages with thousands of characters, such as Japanese and Mandarin, modeling all these characters is problematic due to data scarcity. To alleviate the problem, we propose a multi-task learning model with explicit interaction between characters and syllables by utilizing Self-conditioned connectionist temporal classification (CTC) technique. While the original Self-conditioned CTC estimates character-level intermediate predictions by applying auxiliary CTC losses to a set of intermediate layers, the proposed method additionally estimates syllable-level intermediate predictions in another set of intermediate layers. The character-level and syllable-level predictions are alternately used as conditioning features to deal with mutual dependency between characters and syllables. Experimental results on Japanese and Mandarin datasets show that the proposed multi-sequence intermediate conditioning outperformed the conventional multi-task-based and Self-conditioned CTC-based methods.
Automatic speech recognition (ASR) systems play a key role in many commercial products including voice assistants. Typically, they require large amounts of clean speech data for training which gives an undue advantage to large organizations which have tons of private data. In this paper, we have first curated a fairly big dataset using publicly available data sources. Thereafter, we tried to investigate if we can use publicly available noisy data to train robust ASR systems. We have used speech enhancement to clean the noisy data first and then used it together with its cleaned version to train ASR systems. We have found that using speech enhancement gives 9.5\% better word error rate than training on just noisy data and 9\% better than training on just clean data. It's performance is also comparable to the ideal case scenario when trained on noisy and its clean version.