Quantized Approximate Signal Processing (QASP): Towards Homomorphic Encryption for audio

Audio and speech data are increasingly used in machine learning applications such as speech recognition, speaker identification, and mental health monitoring. However, the passive collection of this data by audio listening devices raises significant privacy concerns. Fully homomorphic encryption (FHE) offers a promising solution by enabling computations on encrypted data and preserving user privacy. Despite its potential, prior attempts to apply FHE to audio processing have faced challenges, particularly in securely computing time frequency representations, a critical step in many audio tasks. Here, we addressed this gap by introducing a fully secure pipeline that computes, with FHE and quantized neural network operations, four fundamental time-frequency representations: Short-Time Fourier Transform (STFT), Mel filterbanks, Mel-frequency cepstral coefficients (MFCCs), and gammatone filters. Our methods also support the private computation of audio descriptors and convolutional neural network (CNN) classifiers. Besides, we proposed approximate STFT algorithms that lighten computation and bit use for statistical and machine learning analyses. We ran experiments on the VocalSet and OxVoc datasets demonstrating the fully private computation of our approach. We showed significant performance improvements with STFT approximation in private statistical analysis of audio markers, and for vocal exercise classification with CNNs. Our results reveal that our approximations substantially reduce error rates compared to conventional STFT implementations in FHE. We also demonstrated a fully private classification based on the raw audio for gender and vocal exercise classification. Finally, we provided a practical heuristic for parameter selection, making quantized approximate signal processing accessible to researchers and practitioners aiming to protect sensitive audio data.

A Comparative Analysis of Static Word Embeddings for Hungarian

This paper presents a comprehensive analysis of various static word embeddings for Hungarian, including traditional models such as Word2Vec, FastText, as well as static embeddings derived from BERT-based models using different extraction methods. We evaluate these embeddings on both intrinsic and extrinsic tasks to provide a holistic view of their performance. For intrinsic evaluation, we employ a word analogy task, which assesses the embeddings ability to capture semantic and syntactic relationships. Our results indicate that traditional static embeddings, particularly FastText, excel in this task, achieving high accuracy and mean reciprocal rank (MRR) scores. Among the BERT-based models, the X2Static method for extracting static embeddings demonstrates superior performance compared to decontextualized and aggregate methods, approaching the effectiveness of traditional static embeddings. For extrinsic evaluation, we utilize a bidirectional LSTM model to perform Named Entity Recognition (NER) and Part-of-Speech (POS) tagging tasks. The results reveal that embeddings derived from dynamic models, especially those extracted using the X2Static method, outperform purely static embeddings. Notably, ELMo embeddings achieve the highest accuracy in both NER and POS tagging tasks, underscoring the benefits of contextualized representations even when used in a static form. Our findings highlight the continued relevance of static word embeddings in NLP applications and the potential of advanced extraction methods to enhance the utility of BERT-based models. This piece of research contributes to the understanding of embedding performance in the Hungarian language and provides valuable insights for future developments in the field. The training scripts, evaluation codes, restricted vocabulary, and extracted embeddings will be made publicly available to support further research and reproducibility.

Empirical Analysis of Asynchronous Federated Learning on Heterogeneous Devices: Efficiency, Fairness, and Privacy Trade-offs

Device heterogeneity poses major challenges in Federated Learning (FL), where resource-constrained clients slow down synchronous schemes that wait for all updates before aggregation. Asynchronous FL addresses this by incorporating updates as they arrive, substantially improving efficiency. While its efficiency gains are well recognized, its privacy costs remain largely unexplored, particularly for high-end devices that contribute updates more frequently, increasing their cumulative privacy exposure. This paper presents the first comprehensive analysis of the efficiency-fairness-privacy trade-off in synchronous vs. asynchronous FL under realistic device heterogeneity. We empirically compare FedAvg and staleness-aware FedAsync using a physical testbed of five edge devices spanning diverse hardware tiers, integrating Local Differential Privacy (LDP) and the Moments Accountant to quantify per-client privacy loss. Using Speech Emotion Recognition (SER) as a privacy-critical benchmark, we show that FedAsync achieves up to 10x faster convergence but exacerbates fairness and privacy disparities: high-end devices contribute 6-10x more updates and incur up to 5x higher privacy loss, while low-end devices suffer amplified accuracy degradation due to infrequent, stale, and noise-perturbed updates. These findings motivate the need for adaptive FL protocols that jointly optimize aggregation and privacy mechanisms based on client capacity and participation dynamics, moving beyond static, one-size-fits-all solutions.

Teochew-Wild: The First In-the-wild Teochew Dataset with Orthographic Annotations

This paper reports the construction of the Teochew-Wild, a speech corpus of the Teochew dialect. The corpus includes 18.9 hours of in-the-wild Teochew speech data from multiple speakers, covering both formal and colloquial expressions, with precise orthographic and pinyin annotations. Additionally, we provide supplementary text processing tools and resources to propel research and applications in speech tasks for this low-resource language, such as automatic speech recognition (ASR) and text-to-speech (TTS). To the best of our knowledge, this is the first publicly available Teochew dataset with accurate orthographic annotations. We conduct experiments on the corpus, and the results validate its effectiveness in ASR and TTS tasks.

Robust Speech Recognition with Schrödinger Bridge-Based Speech Enhancement

In this work, we investigate application of generative speech enhancement to improve the robustness of ASR models in noisy and reverberant conditions. We employ a recently-proposed speech enhancement model based on Schr\"odinger bridge, which has been shown to perform well compared to diffusion-based approaches. We analyze the impact of model scaling and different sampling methods on the ASR performance. Furthermore, we compare the considered model with predictive and diffusion-based baselines and analyze the speech recognition performance when using different pre-trained ASR models. The proposed approach significantly reduces the word error rate, reducing it by approximately 40% relative to the unprocessed speech signals and by approximately 8% relative to a similarly sized predictive approach.

SwinLip: An Efficient Visual Speech Encoder for Lip Reading Using Swin Transformer

This paper presents an efficient visual speech encoder for lip reading. While most recent lip reading studies have been based on the ResNet architecture and have achieved significant success, they are not sufficiently suitable for efficiently capturing lip reading features due to high computational complexity in modeling spatio-temporal information. Additionally, using a complex visual model not only increases the complexity of lip reading models but also induces delays in the overall network for multi-modal studies (e.g., audio-visual speech recognition, speech enhancement, and speech separation). To overcome the limitations of Convolutional Neural Network (CNN)-based models, we apply the hierarchical structure and window self-attention of the Swin Transformer to lip reading. We configure a new lightweight scale of the Swin Transformer suitable for processing lip reading data and present the SwinLip visual speech encoder, which efficiently reduces computational load by integrating modified Convolution-augmented Transformer (Conformer) temporal embeddings with conventional spatial embeddings in the hierarchical structure. Through extensive experiments, we have validated that our SwinLip successfully improves the performance and inference speed of the lip reading network when applied to various backbones for word and sentence recognition, reducing computational load. In particular, our SwinLip demonstrated robust performance in both English LRW and Mandarin LRW-1000 datasets and achieved state-of-the-art performance on the Mandarin LRW-1000 dataset with less computation compared to the existing state-of-the-art model.

Fairness of Automatic Speech Recognition in Cleft Lip and Palate Speech

Speech produced by individuals with cleft lip and palate (CLP) is often highly nasalized and breathy due to structural anomalies, causing shifts in formant structure that affect automatic speech recognition (ASR) performance and fairness. This study hypothesizes that publicly available ASR systems exhibit reduced fairness for CLP speech and confirms this through experiments. Despite formant disruptions, mild and moderate CLP speech retains some spectro-temporal alignment with normal speech, motivating augmentation strategies to enhance fairness. The study systematically explores augmenting CLP speech with normal speech across severity levels and evaluates its impact on ASR fairness. Three ASR models-GMM-HMM, Whisper, and XLS-R-were tested on AIISH and NMCPC datasets. Results indicate that training with normal speech and testing on mixed data improves word error rate (WER). Notably, WER decreased from $22.64\%$ to $18.76\%$ (GMM-HMM, AIISH) and $28.45\%$ to $18.89\%$ (Whisper, NMCPC). The superior performance of GMM-HMM on AIISH may be due to its suitability for Kannada children's speech, a challenge for foundation models like XLS-R and Whisper. To assess fairness, a fairness score was introduced, revealing improvements of $17.89\%$ (AIISH) and $47.50\%$ (NMCPC) with augmentation.

CoGenAV: Versatile Audio-Visual Representation Learning via Contrastive-Generative Synchronization

The inherent synchronization between a speaker's lip movements, voice, and the underlying linguistic content offers a rich source of information for improving speech processing tasks, especially in challenging conditions where traditional audio-only systems falter. We introduce CoGenAV, a powerful and data-efficient model designed to learn versatile audio-visual representations applicable across a wide range of speech and audio-visual tasks. CoGenAV is trained by optimizing a dual objective derived from natural audio-visual synchrony, contrastive feature alignment and generative text prediction, using only 223 hours of labeled data from the LRS2 dataset. This contrastive-generative synchronization strategy effectively captures fundamental cross-modal correlations. We showcase the effectiveness and versatility of the learned CoGenAV representations on multiple benchmarks. When utilized for Audio-Visual Speech Recognition (AVSR) on LRS2, these representations contribute to achieving a state-of-the-art Word Error Rate (WER) of 1.27. They also enable strong performance in Visual Speech Recognition (VSR) with a WER of 22.0 on LRS2, and significantly improve performance in noisy environments by over 70%. Furthermore, CoGenAV representations benefit speech reconstruction tasks, boosting performance in Speech Enhancement and Separation, and achieve competitive results in audio-visual synchronization tasks like Active Speaker Detection (ASD). Our model will be open-sourced to facilitate further development and collaboration within both academia and industry.

SepALM: Audio Language Models Are Error Correctors for Robust Speech Separation

While contemporary speech separation technologies adeptly process lengthy mixed audio waveforms, they are frequently challenged by the intricacies of real-world environments, including noisy and reverberant settings, which can result in artifacts or distortions in the separated speech. To overcome these limitations, we introduce SepALM, a pioneering approach that employs audio language models (ALMs) to rectify and re-synthesize speech within the text domain following preliminary separation. SepALM comprises four core components: a separator, a corrector, a synthesizer, and an aligner. By integrating an ALM-based end-to-end error correction mechanism, we mitigate the risk of error accumulation and circumvent the optimization hurdles typically encountered in conventional methods that amalgamate automatic speech recognition (ASR) with large language models (LLMs). Additionally, we have developed Chain-of-Thought (CoT) prompting and knowledge distillation techniques to facilitate the reasoning and training processes of the ALM. Our experiments substantiate that SepALM not only elevates the precision of speech separation but also markedly bolsters adaptability in novel acoustic environments.

VITA-Audio: Fast Interleaved Cross-Modal Token Generation for Efficient Large Speech-Language Model

With the growing requirement for natural human-computer interaction, speech-based systems receive increasing attention as speech is one of the most common forms of daily communication. However, the existing speech models still experience high latency when generating the first audio token during streaming, which poses a significant bottleneck for deployment. To address this issue, we propose VITA-Audio, an end-to-end large speech model with fast audio-text token generation. Specifically, we introduce a lightweight Multiple Cross-modal Token Prediction (MCTP) module that efficiently generates multiple audio tokens within a single model forward pass, which not only accelerates the inference but also significantly reduces the latency for generating the first audio in streaming scenarios. In addition, a four-stage progressive training strategy is explored to achieve model acceleration with minimal loss of speech quality. To our knowledge, VITA-Audio is the first multi-modal large language model capable of generating audio output during the first forward pass, enabling real-time conversational capabilities with minimal latency. VITA-Audio is fully reproducible and is trained on open-source data only. Experimental results demonstrate that our model achieves an inference speedup of 3~5x at the 7B parameter scale, but also significantly outperforms open-source models of similar model size on multiple benchmarks for automatic speech recognition (ASR), text-to-speech (TTS), and spoken question answering (SQA) tasks.