TidyVoice: A Curated Multilingual Dataset for Speaker Verification Derived from Common Voice

The development of robust, multilingual speaker recognition systems is hindered by a lack of large-scale, publicly available and multilingual datasets, particularly for the read-speech style crucial for applications like anti-spoofing. To address this gap, we introduce the TidyVoice dataset derived from the Mozilla Common Voice corpus after mitigating its inherent speaker heterogeneity within the provided client IDs. TidyVoice currently contains training and test data from over 212,000 monolingual speakers (Tidy-M) and around 4,500 multilingual speakers (Tidy-X) from which we derive two distinct conditions. The Tidy-M condition contains target and non-target trials from monolingual speakers across 81 languages. The Tidy-X condition contains target and non-target trials from multilingual speakers in both same- and cross-language trials. We employ two architectures of ResNet models, achieving a 0.35% EER by fine-tuning on our comprehensive Tidy-M partition. Moreover, we show that this fine-tuning enhances the model's generalization, improving performance on unseen conversational interview data from the CANDOR corpus. The complete dataset, evaluation trials, and our models are publicly released to provide a new resource for the community.

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.

Beyond Mapping : Domain-Invariant Representations via Spectral Embedding of Optimal Transport Plans

Distributional shifts between training and inference time data remain a central challenge in machine learning, often leading to poor performance. It motivated the study of principled approaches for domain alignment, such as optimal transport based unsupervised domain adaptation, that relies on approximating Monge map using transport plans, which is sensitive to the transport problem regularization strategy and hyperparameters, and might yield biased domains alignment. In this work, we propose to interpret smoothed transport plans as adjacency matrices of bipartite graphs connecting source to target domain and derive domain-invariant samples' representations through spectral embedding. We evaluate our approach on acoustic adaptation benchmarks for music genre recognition, music-speech discrimination, as well as electrical cable defect detection and classification tasks using time domain reflection in different diagnosis settings, achieving overall strong performances.

RLBR: Reinforcement Learning with Biasing Rewards for Contextual Speech Large Language Models

Speech large language models (LLMs) have driven significant progress in end-to-end speech understanding and recognition, yet they continue to struggle with accurately recognizing rare words and domain-specific terminology. This paper presents a novel fine-tuning method, Reinforcement Learning with Biasing Rewards (RLBR), which employs a specialized biasing words preferred reward to explicitly emphasize biasing words in the reward calculation. In addition, we introduce reference-aware mechanisms that extend the reinforcement learning algorithm with reference transcription to strengthen the potential trajectory exploration space. Experiments on the LibriSpeech corpus across various biasing list sizes demonstrate that RLBR delivers substantial performance improvements over a strong supervised fine-tuning (SFT) baseline and consistently outperforms several recently published methods. The proposed approach achieves excellent performance on the LibriSpeech test-clean and test-other sets, reaching Biasing Word Error Rates (BWERs) of 0.59% / 2.11%, 1.09% / 3.24%, and 1.36% / 4.04% for biasing list sizes of 100, 500, and 1000, respectively, without compromising the overall WERs.

HoverAI: An Embodied Aerial Agent for Natural Human-Drone Interaction

Drones operating in human-occupied spaces suffer from insufficient communication mechanisms that create uncertainty about their intentions. We present HoverAI, an embodied aerial agent that integrates drone mobility, infrastructure-independent visual projection, and real-time conversational AI into a unified platform. Equipped with a MEMS laser projector, onboard semi-rigid screen, and RGB camera, HoverAI perceives users through vision and voice, responding via lip-synced avatars that adapt appearance to user demographics. The system employs a multimodal pipeline combining VAD, ASR (Whisper), LLM-based intent classification, RAG for dialogue, face analysis for personalization, and voice synthesis (XTTS v2). Evaluation demonstrates high accuracy in command recognition (F1: 0.90), demographic estimation (gender F1: 0.89, age MAE: 5.14 years), and speech transcription (WER: 0.181). By uniting aerial robotics with adaptive conversational AI and self-contained visual output, HoverAI introduces a new class of spatially-aware, socially responsive embodied agents for applications in guidance, assistance, and human-centered interaction.

Unifying Speech Recognition, Synthesis and Conversion with Autoregressive Transformers

Traditional speech systems typically rely on separate, task-specific models for text-to-speech (TTS), automatic speech recognition (ASR), and voice conversion (VC), resulting in fragmented pipelines that limit scalability, efficiency, and cross-task generalization. In this paper, we present General-Purpose Audio (GPA), a unified audio foundation model that integrates multiple core speech tasks within a single large language model (LLM) architecture. GPA operates on a shared discrete audio token space and supports instruction-driven task induction, enabling a single autoregressive model to flexibly perform TTS, ASR, and VC without architectural modifications. This unified design combines a fully autoregressive formulation over discrete speech tokens, joint multi-task training across speech domains, and a scalable inference pipeline that achieves high concurrency and throughput. The resulting model family supports efficient multi-scale deployment, including a lightweight 0.3B-parameter variant optimized for edge and resource-constrained environments. Together, these design choices demonstrate that a unified autoregressive architecture can achieve competitive performance across diverse speech tasks while remaining viable for low-latency, practical deployment.

WenetSpeech-Wu: Datasets, Benchmarks, and Models for a Unified Chinese Wu Dialect Speech Processing Ecosystem

Speech processing for low-resource dialects remains a fundamental challenge in developing inclusive and robust speech technologies. Despite its linguistic significance and large speaker population, the Wu dialect of Chinese has long been hindered by the lack of large-scale speech data, standardized evaluation benchmarks, and publicly available models. In this work, we present WenetSpeech-Wu, the first large-scale, multi-dimensionally annotated open-source speech corpus for the Wu dialect, comprising approximately 8,000 hours of diverse speech data. Building upon this dataset, we introduce WenetSpeech-Wu-Bench, the first standardized and publicly accessible benchmark for systematic evaluation of Wu dialect speech processing, covering automatic speech recognition (ASR), Wu-to-Mandarin translation, speaker attribute prediction, speech emotion recognition, text-to-speech (TTS) synthesis, and instruction-following TTS (instruct TTS). Furthermore, we release a suite of strong open-source models trained on WenetSpeech-Wu, establishing competitive performance across multiple tasks and empirically validating the effectiveness of the proposed dataset. Together, these contributions lay the foundation for a comprehensive Wu dialect speech processing ecosystem, and we open-source proposed datasets, benchmarks, and models to support future research on dialectal speech intelligence.

Speech-Hands: A Self-Reflection Voice Agentic Approach to Speech Recognition and Audio Reasoning with Omni Perception

We introduce a voice-agentic framework that learns one critical omni-understanding skill: knowing when to trust itself versus when to consult external audio perception. Our work is motivated by a crucial yet counterintuitive finding: naively fine-tuning an omni-model on both speech recognition and external sound understanding tasks often degrades performance, as the model can be easily misled by noisy hypotheses. To address this, our framework, Speech-Hands, recasts the problem as an explicit self-reflection decision. This learnable reflection primitive proves effective in preventing the model from being derailed by flawed external candidates. We show that this agentic action mechanism generalizes naturally from speech recognition to complex, multiple-choice audio reasoning. Across the OpenASR leaderboard, Speech-Hands consistently outperforms strong baselines by 12.1% WER on seven benchmarks. The model also achieves 77.37% accuracy and high F1 on audio QA decisions, showing robust generalization and reliability across diverse audio question answering datasets. By unifying perception and decision-making, our work offers a practical path toward more reliable and resilient audio intelligence.

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.

AI-based System for Transforming text and sound to Educational Videos

Technological developments have produced methods that can generate educational videos from input text or sound. Recently, the use of deep learning techniques for image and video generation has been widely explored, particularly in education. However, generating video content from conditional inputs such as text or speech remains a challenging area. In this paper, we introduce a novel method to the educational structure, Generative Adversarial Network (GAN), which develop frame-for-frame frameworks and are able to create full educational videos. The proposed system is structured into three main phases In the first phase, the input (either text or speech) is transcribed using speech recognition. In the second phase, key terms are extracted and relevant images are generated using advanced models such as CLIP and diffusion models to enhance visual quality and semantic alignment. In the final phase, the generated images are synthesized into a video format, integrated with either pre-recorded or synthesized sound, resulting in a fully interactive educational video. The proposed system is compared with other systems such as TGAN, MoCoGAN, and TGANS-C, achieving a Fréchet Inception Distance (FID) score of 28.75%, which indicates improved visual quality and better over existing methods.