Linhao Dong

Minglun Han, Ye Bai, Chen Shen et al.

Speech self-supervised pre-training can effectively improve the performance of downstream tasks. However, previous self-supervised learning (SSL) methods for speech, such as HuBERT and BEST-RQ, focus on utilizing non-causal encoders with bidirectional context, and lack sufficient support for downstream streaming models. To address this issue, we introduce the next token prediction based speech pre-training method with random-projection quantizer (NEST-RQ). NEST-RQ employs causal encoders with only left context and uses next token prediction (NTP) as the training task. On the large-scale dataset, compared to BEST-RQ, the proposed NEST-RQ achieves comparable performance on non-streaming automatic speech recognition (ASR) and better performance on streaming ASR. We also conduct analytical experiments in terms of the future context size of streaming ASR, the codebook quality of SSL and the model size of the encoder. In summary, the paper demonstrates the feasibility of the NTP in speech SSL and provides empirical evidence and insights for speech SSL research.

Qijie Shao, Linhao Dong, Kun Wei et al.

Data2vec is a self-supervised learning (SSL) approach that employs a teacher-student architecture for contextual representation learning via masked prediction, demonstrating remarkable performance in monolingual ASR. Previous studies have revealed that data2vec's shallow layers capture speaker and language information, middle layers encode phoneme and word features, while deep layers are responsible for reconstruction. Language and phoneme features are crucial for multilingual ASR. However, data2vec's masked representation generation relies on multi-layer averaging, inevitably coupling these features. To address this limitation, we propose a decoupling quantization based data2vec (DQ-Data2vec) for multilingual ASR, which includes a data2vec backbone and two improved online K-means quantizers. Our core idea is using the K-means quantizer with specified cluster numbers to decouple language and phoneme information for masked prediction. Specifically, in the language quantization, considering that the number of languages is significantly different from other irrelevant features (e.g., speakers), we assign the cluster number to match the number of languages, explicitly decoupling shallow layers' language-related information from irrelevant features. This strategy is also applied to decoupling middle layers' phoneme and word features. In a self-supervised scenario, experiments on the CommonVoice dataset demonstrate that DQ-Data2vec achieves a relative reduction of 9.51% in phoneme error rate (PER) and 11.58% in word error rate (WER) compared to data2vec and UniData2vec. Moreover, in a weakly-supervised scenario incorporating language labels and high-resource language text labels, the relative reduction is 18.09% and 1.55%, respectively.

Linhao Dong, Zhecheng An, Peihao Wu et al.

Speech or text representation generated by pre-trained models contains modal-specific information that could be combined for benefiting spoken language understanding (SLU) tasks. In this work, we propose a novel pre-training paradigm termed Continuous Integrate-and-Fire Pre-Training (CIF-PT). It relies on a simple but effective frame-to-token alignment: continuous integrate-and-fire (CIF) to bridge the representations between speech and text. It jointly performs speech-to-text training and language model distillation through CIF as the pre-training (PT). Evaluated on SLU benchmark SLURP dataset, CIF-PT outperforms the state-of-the-art model by 1.94% of accuracy and 2.71% of SLU-F1 on the tasks of intent classification and slot filling, respectively. We also observe the cross-modal representation extracted by CIF-PT obtains better performance than other neural interfaces for the tasks of SLU, including the dominant speech representation learned from self-supervised pre-training.

Minglun Han, Linhao Dong, Zhenlin Liang et al.

Nowadays, most methods in end-to-end contextual speech recognition bias the recognition process towards contextual knowledge. Since all-neural contextual biasing methods rely on phrase-level contextual modeling and attention-based relevance modeling, they may encounter confusion between similar context-specific phrases, which hurts predictions at the token level. In this work, we focus on mitigating confusion problems with fine-grained contextual knowledge selection (FineCoS). In FineCoS, we introduce fine-grained knowledge to reduce the uncertainty of token predictions. Specifically, we first apply phrase selection to narrow the range of phrase candidates, and then conduct token attention on the tokens in the selected phrase candidates. Moreover, we re-normalize the attention weights of most relevant phrases in inference to obtain more focused phrase-level contextual representations, and inject position information to better discriminate phrases or tokens. On LibriSpeech and an in-house 160,000-hour dataset, we explore the proposed methods based on a controllable all-neural biasing method, collaborative decoding (ColDec). The proposed methods provide at most 6.1% relative word error rate reduction on LibriSpeech and 16.4% relative character error rate reduction on the in-house dataset over ColDec.

Minglun Han, Linhao Dong, Shiyu Zhou et al.

End-to-end (E2E) models have achieved promising results on multiple speech recognition benchmarks, and shown the potential to become the mainstream. However, the unified structure and the E2E training hamper injecting contextual information into them for contextual biasing. Though contextual LAS (CLAS) gives an excellent all-neural solution, the degree of biasing to given context information is not explicitly controllable. In this paper, we focus on incorporating context information into the continuous integrate-and-fire (CIF) based model that supports contextual biasing in a more controllable fashion. Specifically, an extra context processing network is introduced to extract contextual embeddings, integrate acoustically relevant context information and decode the contextual output distribution, thus forming a collaborative decoding with the decoder of the CIF-based model. Evaluated on the named entity rich evaluation sets of HKUST/AISHELL-2, our method brings relative character error rate (CER) reduction of 8.83%/21.13% and relative named entity character error rate (NE-CER) reduction of 40.14%/51.50% when compared with a strong baseline. Besides, it keeps the performance on original evaluation set without degradation.

Linhao Dong, Cheng Yi, Jianzong Wang et al.

End-to-end models are gaining wider attention in the field of automatic speech recognition (ASR). One of their advantages is the simplicity of building that directly recognizes the speech frame sequence into the text label sequence by neural networks. According to the driving end in the recognition process, end-to-end ASR models could be categorized into two types: label-synchronous and frame-synchronous, each of which has unique model behaviour and characteristic. In this work, we make a detailed comparison on a representative label-synchronous model (transformer) and a soft frame-synchronous model (continuous integrate-and-fire (CIF) based model). The results on three public dataset and a large-scale dataset with 12000 hours of training data show that the two types of models have respective advantages that are consistent with their synchronous mode.

Linhao Dong, Bo Xu

In this paper, we propose a novel soft and monotonic alignment mechanism used for sequence transduction. It is inspired by the integrate-and-fire model in spiking neural networks and employed in the encoder-decoder framework consists of continuous functions, thus being named as: Continuous Integrate-and-Fire (CIF). Applied to the ASR task, CIF not only shows a concise calculation, but also supports online recognition and acoustic boundary positioning, thus suitable for various ASR scenarios. Several support strategies are also proposed to alleviate the unique problems of CIF-based model. With the joint action of these methods, the CIF-based model shows competitive performance. Notably, it achieves a word error rate (WER) of 2.86% on the test-clean of Librispeech and creates new state-of-the-art result on Mandarin telephone ASR benchmark.

Linhao Dong, Feng Wang, Bo Xu

Self-attention network, an attention-based feedforward neural network, has recently shown the potential to replace recurrent neural networks (RNNs) in a variety of NLP tasks. However, it is not clear if the self-attention network could be a good alternative of RNNs in automatic speech recognition (ASR), which processes the longer speech sequences and may have online recognition requirements. In this paper, we present a RNN-free end-to-end model: self-attention aligner (SAA), which applies the self-attention networks to a simplified recurrent neural aligner (RNA) framework. We also propose a chunk-hopping mechanism, which enables the SAA model to encode on segmented frame chunks one after another to support online recognition. Experiments on two Mandarin ASR datasets show the replacement of RNNs by the self-attention networks yields a 8.4%-10.2% relative character error rate (CER) reduction. In addition, the chunk-hopping mechanism allows the SAA to have only a 2.5% relative CER degradation with a 320ms latency. After jointly training with a self-attention network language model, our SAA model obtains further error rate reduction on multiple datasets. Especially, it achieves 24.12% CER on the Mandarin ASR benchmark (HKUST), exceeding the best end-to-end model by over 2% absolute CER.

Linhao Dong, Shiyu Zhou, Wei Chen et al.

End-to-end models have been showing superiority in Automatic Speech Recognition (ASR). At the same time, the capacity of streaming recognition has become a growing requirement for end-to-end models. Following these trends, an encoder-decoder recurrent neural network called Recurrent Neural Aligner (RNA) has been freshly proposed and shown its competitiveness on two English ASR tasks. However, it is not clear if RNA can be further improved and applied to other spoken language. In this work, we explore the applicability of RNA in Mandarin Chinese and present four effective extensions: In the encoder, we redesign the temporal down-sampling and introduce a powerful convolutional structure. In the decoder, we utilize a regularizer to smooth the output distribution and conduct joint training with a language model. On two Mandarin Chinese conversational telephone speech recognition (MTS) datasets, our Extended-RNA obtains promising performance. Particularly, it achieves 27.7% character error rate (CER), which is superior to current state-of-the-art result on the popular HKUST task.

Shiyu Zhou, Linhao Dong, Shuang Xu et al.

The choice of modeling units is critical to automatic speech recognition (ASR) tasks. Conventional ASR systems typically choose context-dependent states (CD-states) or context-dependent phonemes (CD-phonemes) as their modeling units. However, it has been challenged by sequence-to-sequence attention-based models, which integrate an acoustic, pronunciation and language model into a single neural network. On English ASR tasks, previous attempts have already shown that the modeling unit of graphemes can outperform that of phonemes by sequence-to-sequence attention-based model. In this paper, we are concerned with modeling units on Mandarin Chinese ASR tasks using sequence-to-sequence attention-based models with the Transformer. Five modeling units are explored including context-independent phonemes (CI-phonemes), syllables, words, sub-words and characters. Experiments on HKUST datasets demonstrate that the lexicon free modeling units can outperform lexicon related modeling units in terms of character error rate (CER). Among five modeling units, character based model performs best and establishes a new state-of-the-art CER of $26.64\%$ on HKUST datasets without a hand-designed lexicon and an extra language model integration, which corresponds to a $4.8\%$ relative improvement over the existing best CER of $28.0\%$ by the joint CTC-attention based encoder-decoder network.

Shiyu Zhou, Linhao Dong, Shuang Xu et al.

Sequence-to-sequence attention-based models have recently shown very promising results on automatic speech recognition (ASR) tasks, which integrate an acoustic, pronunciation and language model into a single neural network. In these models, the Transformer, a new sequence-to-sequence attention-based model relying entirely on self-attention without using RNNs or convolutions, achieves a new single-model state-of-the-art BLEU on neural machine translation (NMT) tasks. Since the outstanding performance of the Transformer, we extend it to speech and concentrate on it as the basic architecture of sequence-to-sequence attention-based model on Mandarin Chinese ASR tasks. Furthermore, we investigate a comparison between syllable based model and context-independent phoneme (CI-phoneme) based model with the Transformer in Mandarin Chinese. Additionally, a greedy cascading decoder with the Transformer is proposed for mapping CI-phoneme sequences and syllable sequences into word sequences. Experiments on HKUST datasets demonstrate that syllable based model with the Transformer performs better than CI-phoneme based counterpart, and achieves a character error rate (CER) of \emph{$28.77\%$}, which is competitive to the state-of-the-art CER of $28.0\%$ by the joint CTC-attention based encoder-decoder network.