In speaker-independent speech emotion recognition, the training and testing samples are collected from diverse speakers, leading to a multi-domain shift challenge across the feature distributions of data from different speakers. Consequently, when the trained model is confronted with data from new speakers, its performance tends to degrade. To address the issue, we propose a Dynamic Joint Distribution Adaptation (DJDA) method under the framework of multi-source domain adaptation. DJDA firstly utilizes joint distribution adaptation (JDA), involving marginal distribution adaptation (MDA) and conditional distribution adaptation (CDA), to more precisely measure the multi-domain distribution shifts caused by different speakers. This helps eliminate speaker bias in emotion features, allowing for learning discriminative and speaker-invariant speech emotion features from coarse-level to fine-level. Furthermore, we quantify the adaptation contributions of MDA and CDA within JDA by using a dynamic balance factor based on $\mathcal{A}$-Distance, promoting to effectively handle the unknown distributions encountered in data from new speakers. Experimental results demonstrate the superior performance of our DJDA as compared to other state-of-the-art (SOTA) methods.
Recently audio-visual speech recognition (AVSR), which better leverages video modality as additional information to extend automatic speech recognition (ASR), has shown promising results in complex acoustic environments. However, there is still substantial space to improve as complex computation of visual modules and ineffective fusion of audio-visual modalities. To eliminate these drawbacks, we propose a down-up sampling-based AVSR model (Hourglass-AVSR) to enjoy high efficiency and performance, whose time length is scaled during the intermediate processing, resembling an hourglass. Firstly, we propose a context and residual aware video upsampling approach to improve the recognition performance, which utilizes contextual information from visual representations and captures residual information between adjacent video frames. Secondly, we introduce a visual-audio alignment approach during the upsampling by explicitly incorporating boundary constraint loss. Besides, we propose a cross-layer attention fusion to capture the modality dependencies within each visual encoder layer. Experiments conducted on the MISP-AVSR dataset reveal that our proposed Hourglass-AVSR model outperforms ASR model by 12.9% and 20.8% relative concatenated minimum permutation character error rate (cpCER) reduction on far-field and middle-field test sets, respectively. Moreover, compared to other state-of-the-art AVSR models, our model exhibits the highest improvement in cpCER for the visual module. Furthermore, on the benefit of our down-up sampling approach, Hourglass-AVSR model reduces 54.2% overall computation costs with minor performance degradation.
Unsupervised learning objectives like language modeling and de-noising constitute a significant part in producing pre-trained models that perform various downstream applications from natural language understanding to conversational tasks. However, despite impressive generative capabilities of recent large language models, their abilities to capture syntactic or semantic structure within text lag behind. We hypothesize that the mismatch between linguistic performance and competence in machines is attributable to insufficient transfer of linguistic structure knowledge to computational systems with currently popular pre-training objectives. We show that punctuation restoration as a learning objective improves in- and out-of-distribution performance on structure-related tasks like named entity recognition, open information extraction, chunking, and part-of-speech tagging. Punctuation restoration is an effective learning objective that can improve structure understanding and yield a more robust structure-aware representations of natural language.
Continued self-supervised (SSL) pre-training for adapting existing SSL models to the target domain has shown to be extremely effective for low-resource Automatic Speech Recognition (ASR). This paper proposes Stable Distillation, a simple and novel approach for SSL-based continued pre-training that boosts ASR performance in the target domain where both labeled and unlabeled data are limited. Stable Distillation employs self-distillation as regularization for continued pre-training, alleviating the over-fitting issue, a common problem continued pre-training faces when the source and target domains differ. Specifically, first, we perform vanilla continued pre-training on an initial SSL pre-trained model on the target domain ASR dataset and call it the teacher. Next, we take the same initial pre-trained model as a student to perform continued pre-training while enforcing its hidden representations to be close to that of the teacher (via MSE loss). This student is then used for downstream ASR fine-tuning on the target dataset. In practice, Stable Distillation outperforms all our baselines by 0.8 - 7 WER when evaluated in various experimental settings.
The widespread smart devices raise people's concerns of being eavesdropped on. To enhance voice privacy, recent studies exploit the nonlinearity in microphone to jam audio recorders with inaudible ultrasound. However, existing solutions solely rely on energetic masking. Their simple-form noise leads to several problems, such as high energy requirements and being easily removed by speech enhancement techniques. Besides, most of these solutions do not support authorized recording, which restricts their usage scenarios. In this paper, we design an efficient yet robust system that can jam microphones while preserving authorized recording. Specifically, we propose a novel phoneme-based noise with the idea of informational masking, which can distract both machines and humans and is resistant to denoising techniques. Besides, we optimize the noise transmission strategy for broader coverage and implement a hardware prototype of our system. Experimental results show that our system can reduce the recognition accuracy of recordings to below 50\% under all tested speech recognition systems, which is much better than existing solutions.
The Streaming Unmixing and Recognition Transducer (SURT) has recently become a popular framework for continuous, streaming, multi-talker speech recognition (ASR). With advances in architecture, objectives, and mixture simulation methods, it was demonstrated that SURT can be an efficient streaming method for speaker-agnostic transcription of real meetings. In this work, we push this framework further by proposing methods to perform speaker-attributed transcription with SURT, for both short mixtures and long recordings. We achieve this by adding an auxiliary speaker branch to SURT, and synchronizing its label prediction with ASR token prediction through HAT-style blank factorization. In order to ensure consistency in relative speaker labels across different utterance groups in a recording, we propose "speaker prefixing" -- appending each chunk with high-confidence frames of speakers identified in previous chunks, to establish the relative order. We perform extensive ablation experiments on synthetic LibriSpeech mixtures to validate our design choices, and demonstrate the efficacy of our final model on the AMI corpus.
Large language models (LLMs) have significantly advanced Natural Language Processing (NLP) tasks in recent years. However, their universal nature poses limitations in scenarios requiring personalized responses, such as recommendation systems and chatbots. This paper investigates methods to personalize LLMs, comparing fine-tuning and zero-shot reasoning approaches on subjective tasks. Results demonstrate that personalized fine-tuning improves model reasoning compared to non-personalized models. Experiments on datasets for emotion recognition and hate speech detection show consistent performance gains with personalized methods across different LLM architectures. These findings underscore the importance of personalization for enhancing LLM capabilities in subjective text perception tasks.
Speech-to-Text Translation (S2TT) has typically been addressed with cascade systems, where speech recognition systems generate a transcription that is subsequently passed to a translation model. While there has been a growing interest in developing direct speech translation systems to avoid propagating errors and losing non-verbal content, prior work in direct S2TT has struggled to conclusively establish the advantages of integrating the acoustic signal directly into the translation process. This work proposes using contrastive evaluation to quantitatively measure the ability of direct S2TT systems to disambiguate utterances where prosody plays a crucial role. Specifically, we evaluated Korean-English translation systems on a test set containing wh-phrases, for which prosodic features are necessary to produce translations with the correct intent, whether it's a statement, a yes/no question, a wh-question, and more. Our results clearly demonstrate the value of direct translation systems over cascade translation models, with a notable 12.9% improvement in overall accuracy in ambiguous cases, along with up to a 15.6% increase in F1 scores for one of the major intent categories. To the best of our knowledge, this work stands as the first to provide quantitative evidence that direct S2TT models can effectively leverage prosody. The code for our evaluation is openly accessible and freely available for review and utilisation.
Self-supervised speech pre-training methods have developed rapidly in recent years, which show to be very effective for many near-field single-channel speech tasks. However, far-field multichannel speech processing is suffering from the scarcity of labeled multichannel data and complex ambient noises. The efficacy of self-supervised learning for far-field multichannel and multi-modal speech processing has not been well explored. Considering that visual information helps to improve speech recognition performance in noisy scenes, in this work we propose a multichannel multi-modal speech self-supervised learning framework AV-wav2vec2, which utilizes video and multichannel audio data as inputs. First, we propose a multi-path structure to process multichannel audio streams and a visual stream in parallel, with intra- and inter-channel contrastive losses as training targets to fully exploit the spatiotemporal information in multichannel speech data. Second, based on contrastive learning, we use additional single-channel audio data, which is trained jointly to improve the performance of speech representation. Finally, we use a Chinese multichannel multi-modal dataset in real scenarios to validate the effectiveness of the proposed method on audio-visual speech recognition (AVSR), automatic speech recognition (ASR), visual speech recognition (VSR) and audio-visual speaker diarization (AVSD) tasks.
Neural networks have become popular due to their versatility and state-of-the-art results in many applications, such as image classification, natural language processing, speech recognition, forecasting, etc. These applications are also used in resource-constrained environments such as embedded devices. In this work, the susceptibility of neural network implementations to reverse engineering is explored on the NVIDIA Jetson Nano microcomputer via side-channel analysis. To this end, an architecture extraction attack is presented. In the attack, 15 popular convolutional neural network architectures (EfficientNets, MobileNets, NasNet, etc.) are implemented on the GPU of Jetson Nano and the electromagnetic radiation of the GPU is analyzed during the inference operation of the neural networks. The results of the analysis show that neural network architectures are easily distinguishable using deep learning-based side-channel analysis.