The study investigates the limitations of Phonetic Foundation Models (PFMs) for Speech-to-IPA transcription using Grapheme-to-Phoneme (G2P) labels and proposes a new approach based on continuous Artic…

The paper proposes a novel workflow to extract fine-grained regional accent features in Brazilian Portuguese using only acoustic labels and a phoneme-based forced aligner, showing that localized featu…

This paper investigates the use of various acoustic features for recognizing dysarthric speech using a Factorized Time Delay Neural Network (F-TDNN) model, achieving a relative improvement of 4.65% in…

The paper repurposes a pre-trained speech classifier as the backbone for diffusion generation, reducing the need for two separately trained models.

The paper introduces SURE, a unified framework designed to standardize and improve the comparability and reproducibility of evaluations for advanced speech understanding models.

This paper proposes a model for few-shot class-incremental audio classification, which consists of a pseudo-incrementally trained embedding learner and a continually updated stochastic classifier.

The authors propose a method for phone segmentation and recognition using self-supervised speech models, requiring minimal phonetic transcriptions and generalizing to unseen phones.

This paper investigates the effectiveness of incorporating synthetic speech data in Automatic Speech Recognition (ASR) Systems for three Indic languages by analyzing performance gains, script sources,…

PolySpeech-100 introduces a massive, multi-lingual benchmark covering 110 linguistic variants to rigorously test Speech-LLMs, demonstrating that open-source models struggle with low-resource languages…

The paper introduces GigaSpeechBench, a comprehensive multilingual and multidimensional ASR & AST benchmark with 680 hours of human-annotated speech, featuring 12 low-resource languages, 6 Chinese dia…

The paper proposes a multiple instance neural network architecture using fine-tuned wav2vec 2.0, WavLM and Whisper encoders for stuttering detection and classification, achieving improvements in both…

The paper proposes SpeechKV, a method to compress speech sequences inside large language models using a learned pooling, maintaining performance and delivering decoding speedup.

This paper investigates the correlation between dimensions of self-supervised speech features and speech characteristics, finding that content dimensions primarily capture intensity, formants, and voi…

The authors present ZONOS2 8B, a TTS model with improved naturalness, prosody, and voice cloning fidelity, achieved through scaling, data expansion, and simplification.

This paper proposes an ensemble of two source separators, FTRNN and TF-Locoformer, trained with mixture invariant training (MixIT), and introduces mixture-constrained max pooling (MCM) to improve bird…

This paper explores the effect of conversational timing properties on automatic speech recognition (ASR) systems by controlling and optimizing pause and overlap timing distributions.

This paper proposes a graph framework using pretrained SSL embeddings for speech analysis in Amyotrophic Lateral Sclerosis (ALS) patients, achieving better results than validation baselines on the SAN…

The paper proposes SALMONN-2, an ALLM built on a unified SSL encoder, and presents a multi-layer feature fusion adapter to better exploit hierarchical SSL encoder representations. It also explores mul…

This paper proposes a multimodal framework for jointly improving Automatic Speech Recognition (ASR) and Dialect Identification (DID) in Indian languages using a Bottleneck Encoder, RoBERTa encoder, ga…