Bone-conduction Guided Multimodal Speech Enhancement with Conditional Diffusion Models

Single-channel speech enhancement models face significant performance degradation in extremely noisy environments. While prior work has shown that complementary bone-conducted speech can guide enhancement, effective integration of this noise-immune modality remains a challenge. This paper introduces a novel multimodal speech enhancement framework that integrates bone-conduction sensors with air-conducted microphones using a conditional diffusion model. Our proposed model significantly outperforms previously established multimodal techniques and a powerful diffusion-based single-modal baseline across a wide range of acoustic conditions.

A Unified Neural Codec Language Model for Selective Editable Text to Speech Generation

Neural codec language models achieve impressive zero-shot Text-to-Speech (TTS) by fully imitating the acoustic characteristics of a short speech prompt, including timbre, prosody, and paralinguistic information. However, such holistic imitation limits their ability to isolate and control individual attributes. In this paper, we present a unified codec language model SpeechEdit that extends zero-shot TTS with a selective control mechanism. By default, SpeechEdit reproduces the complete acoustic profile inferred from the speech prompt, but it selectively overrides only the attributes specified by explicit control instructions. To enable controllable modeling, SpeechEdit is trained on our newly constructed LibriEdit dataset, which provides delta (difference-aware) training pairs derived from LibriHeavy. Experimental results show that our approach maintains naturalness and robustness while offering flexible and localized control over desired attributes. Audio samples are available at https://speech-editing.github.io/speech-editing/.

ParaMETA: Towards Learning Disentangled Paralinguistic Speaking Styles Representations from Speech

Learning representative embeddings for different types of speaking styles, such as emotion, age, and gender, is critical for both recognition tasks (e.g., cognitive computing and human-computer interaction) and generative tasks (e.g., style-controllable speech generation). In this work, we introduce ParaMETA, a unified and flexible framework for learning and controlling speaking styles directly from speech. Unlike existing methods that rely on single-task models or cross-modal alignment, ParaMETA learns disentangled, task-specific embeddings by projecting speech into dedicated subspaces for each type of style. This design reduces inter-task interference, mitigates negative transfer, and allows a single model to handle multiple paralinguistic tasks such as emotion, gender, age, and language classification. Beyond recognition, ParaMETA enables fine-grained style control in Text-To-Speech (TTS) generative models. It supports both speech- and text-based prompting and allows users to modify one speaking styles while preserving others. Extensive experiments demonstrate that ParaMETA outperforms strong baselines in classification accuracy and generates more natural and expressive speech, while maintaining a lightweight and efficient model suitable for real-world applications.

SSVD-O: Parameter-Efficient Fine-Tuning with Structured SVD for Speech Recognition

Parameter-efficient fine-tuning (PEFT) is a scalable approach for adapting large speech foundation models to new domains. While methods such as LoRA and its state-of-the-art variants reduce adaptation costs, they typically allocate parameters uniformly across model subspaces, which limits their efficiency and scalability in speech applications. Building on our prior work, this paper introduces SSVD-Outer (SSVD-O), an extension of the structured SVD-guided (SSVD) fine-tuning method. SSVD-O combines input acoustic feature space-associated inner transformations with output semantic feature space-associated outer transformations to enable scalable and balanced adaptation. We conduct the first systematic analysis of parameter budget allocation across model subspaces in PEFT for automatic speech recognition (ASR), and investigate the trade-off between learning and forgetting under constrained resources. SSVD-O is benchmarked against LoRA, DoRA, PiSSA, and SSVD on domain-shifted ASR tasks, including child speech and regional accents, across model scales from 0.1B to 2B within the ESPnet framework. Experimental results show that SSVD-O consistently narrows the performance gap to full fine-tuning while improving generalization and mitigating catastrophic forgetting.

CTC-DID: CTC-Based Arabic dialect identification for streaming applications

This paper proposes a Dialect Identification (DID) approach inspired by the Connectionist Temporal Classification (CTC) loss function as used in Automatic Speech Recognition (ASR). CTC-DID frames the dialect identification task as a limited-vocabulary ASR system, where dialect tags are treated as a sequence of labels for a given utterance. For training, the repetition of dialect tags in transcriptions is estimated either using a proposed Language-Agnostic Heuristic (LAH) approach or a pre-trained ASR model. The method is evaluated on the low-resource Arabic Dialect Identification (ADI) task, with experimental results demonstrating that an SSL-based CTC-DID model, trained on a limited dataset, outperforms both fine-tuned Whisper and ECAPA-TDNN models. Notably, CTC-DID also surpasses these models in zero-shot evaluation on the Casablanca dataset. The proposed approach is found to be more robust to shorter utterances and is shown to be easily adaptable for streaming, real-time applications, with minimal performance degradation.

Do Neural Codecs Generalize? A Controlled Study Across Unseen Languages and Non-Speech Tasks

This paper investigates three crucial yet underexplored aspects of the generalization capabilities of neural audio codecs (NACs): (i) whether NACs can generalize to unseen languages during pre-training, (ii) whether speech-only pre-trained NACs can effectively generalize to non-speech applications such as environmental sounds, music, and animal vocalizations, and (iii) whether incorporating non-speech data during pre-training can improve performance on both speech and non-speech tasks. Existing studies typically rely on off-the-shelf NACs for comparison, which limits insight due to variations in implementation. In this work, we train NACs from scratch using strictly controlled configurations and carefully curated pre-training data to enable fair comparisons. We conduct a comprehensive evaluation of NAC performance on both signal reconstruction quality and downstream applications using 11 metrics. Our results show that NACs can generalize to unseen languages during pre-training, speech-only pre-trained NACs exhibit degraded performance on non-speech tasks, and incorporating non-speech data during pre-training improves performance on non-speech tasks while maintaining comparable performance on speech tasks.

Purification Before Fusion: Toward Mask-Free Speech Enhancement for Robust Audio-Visual Speech Recognition

Audio-visual speech recognition (AVSR) typically improves recognition accuracy in noisy environments by integrating noise-immune visual cues with audio signals. Nevertheless, high-noise audio inputs are prone to introducing adverse interference into the feature fusion process. To mitigate this, recent AVSR methods often adopt mask-based strategies to filter audio noise during feature interaction and fusion, yet such methods risk discarding semantically relevant information alongside noise. In this work, we propose an end-to-end noise-robust AVSR framework coupled with speech enhancement, eliminating the need for explicit noise mask generation. This framework leverages a Conformer-based bottleneck fusion module to implicitly refine noisy audio features with video assistance. By reducing modality redundancy and enhancing inter-modal interactions, our method preserves speech semantic integrity to achieve robust recognition performance. Experimental evaluations on the public LRS3 benchmark suggest that our method outperforms prior advanced mask-based baselines under noisy conditions.

Robust Online Overdetermined Independent Vector Analysis Based on Bilinear Decomposition

Online blind source separation is essential for both speech communication and human-machine interaction. Among existing approaches, overdetermined independent vector analysis (OverIVA) delivers strong performance by exploiting the statistical independence of source signals and the orthogonality between source and noise subspaces. However, when applied to large microphone arrays, the number of parameters grows rapidly, which can degrade online estimation accuracy. To overcome this challenge, we propose decomposing each long separation filter into a bilinear form of two shorter filters, thereby reducing the number of parameters. Because the two filters are closely coupled, we design an alternating iterative projection algorithm to update them in turn. Simulation results show that, with far fewer parameters, the proposed method achieves improved performance and robustness.

Song Aesthetics Evaluation with Multi-Stem Attention and Hierarchical Uncertainty Modeling

Music generative artificial intelligence (AI) is rapidly expanding music content, necessitating automated song aesthetics evaluation. However, existing studies largely focus on speech, audio or singing quality, leaving song aesthetics underexplored. Moreover, conventional approaches often predict a precise Mean Opinion Score (MOS) value directly, which struggles to capture the nuances of human perception in song aesthetics evaluation. This paper proposes a song-oriented aesthetics evaluation framework, featuring two novel modules: 1) Multi-Stem Attention Fusion (MSAF) builds bidirectional cross-attention between mixture-vocal and mixture-accompaniment pairs, fusing them to capture complex musical features; 2) Hierarchical Granularity-Aware Interval Aggregation (HiGIA) learns multi-granularity score probability distributions, aggregates them into a score interval, and applies a regression within the interval to produce the final score. We evaluated on two datasets of full-length songs: SongEval dataset (AI-generated) and an internal aesthetics dataset (human-created), and compared with two state-of-the-art (SOTA) models. Results show that the proposed method achieves stronger performance for multi-dimensional song aesthetics evaluation.

Confidence-based Filtering for Speech Dataset Curation with Generative Speech Enhancement Using Discrete Tokens

Generative speech enhancement (GSE) models show great promise in producing high-quality clean speech from noisy inputs, enabling applications such as curating noisy text-to-speech (TTS) datasets into high-quality ones. However, GSE models are prone to hallucination errors, such as phoneme omissions and speaker inconsistency, which conventional error filtering based on non-intrusive speech quality metrics often fails to detect. To address this issue, we propose a non-intrusive method for filtering hallucination errors from discrete token-based GSE models. Our method leverages the log-probabilities of generated tokens as confidence scores to detect potential errors. Experimental results show that the confidence scores strongly correlate with a suite of intrusive SE metrics, and that our method effectively identifies hallucination errors missed by conventional filtering methods. Furthermore, we demonstrate the practical utility of our method: curating an in-the-wild TTS dataset with our confidence-based filtering improves the performance of subsequently trained TTS models.