Diffusion-based text-to-speech (TTS) systems have made remarkable progress in zero-shot speech synthesis, yet optimizing all components for perceptual metrics remains challenging. Prior work with DMOSpeech demonstrated direct metric optimization for speech generation components, but duration prediction remained unoptimized. This paper presents DMOSpeech 2, which extends metric optimization to the duration predictor through a reinforcement learning approach. The proposed system implements a novel duration policy framework using group relative preference optimization (GRPO) with speaker similarity and word error rate as reward signals. By optimizing this previously unoptimized component, DMOSpeech 2 creates a more complete metric-optimized synthesis pipeline. Additionally, this paper introduces teacher-guided sampling, a hybrid approach leveraging a teacher model for initial denoising steps before transitioning to the student model, significantly improving output diversity while maintaining efficiency. Comprehensive evaluations demonstrate superior performance across all metrics compared to previous systems, while reducing sampling steps by half without quality degradation. These advances represent a significant step toward speech synthesis systems with metric optimization across multiple components.

Xilin Jiang, Sukru Samet Dindar, Vishal Choudhari, Stephan Bickel, Ashesh Mehta, Guy M McKhann, Daniel Friedman, Adeen Flinker, Nima Mesgarani. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Transformer-based speech language models (SLMs) have significantly improved neural speech recognition and understanding.While existing research has examined how well SLMs encode shallow acoustic and phonetic features, the extent to which SLMs encode nuanced syntactic and conceptual features remains unclear.By drawing parallels with linguistic competence assessments for large language models, this study is the first to systematically evaluate the presence of contextual syntactic and semantic features across SLMs for self-supervised learning (S3M), automatic speech recognition (ASR), speech compression (codec), and as the encoder for auditory large language models (AudioLLMs).Through minimal pair designs and diagnostic feature analysis across 71 tasks spanning diverse linguistic levels, our layer-wise and time-resolved analysis uncovers that 1) all speech encode grammatical features more robustly than conceptual ones.2) Despite never seeing text, S3M match or surpass ASR encoders on every linguistic level, demonstrating that rich grammatical and even conceptual knowledge can arise purely from audio.3) S3M representations peak mid-network and then crash in the final layers, whereas ASR and AudioLLM encoders maintain or improve, reflecting how pre-training objectives reshape late-layer content.4) Temporal probing further shows that S3Ms encode grammatical cues 500 ms before a word begins, whereas AudioLLMs distribute evidence more evenly-indicating that objectives shape not only where but also when linguistic information is most salient.Together, these findings establish the first largescale map of contextual syntax and semantics in speech models and highlight both the promise and the limits of current SLM training paradigms. * Equal contribution.

Transformers have been the most successful architecture for various speech modeling tasks, including speech separation. However, the self-attention mechanism in transformers with quadratic complexity is inefficient in computation and memory. Recent models incorporate new layers and modules along with transformers for better performance but also introduce extra model complexity. In this work, we replace transformers with Mamba, a selective state space model, for speech separation. We propose dual-path Mamba, which models short-term and long-term forward and backward dependency of speech signals using selective state spaces. Our experimental results on the WSJ0-2mix data show that our dual-path Mamba models of comparably smaller sizes outperform state-of-the-art RNN model DPRNN, CNN model WaveSplit, and transformer model Sepformer. Our largest model reaches an SI-SNRi of 23.4 dB. Code available: https://github.com/xi-j/Mamba-TasNet

It is too early to conclude that Mamba is a better alternative to transformers for speech before comparing Mamba with transformers in terms of both performance and efficiency in multiple speech-related tasks. To reach this conclusion, we propose and evaluate three models for three tasks: Mamba-TasNet for speech separation, ConMamba for speech recognition, and VALL-M for speech synthesis. We compare them with transformers of similar sizes in performance, memory, and speed. Our Mamba or Mamba-transformer hybrid models show comparable or higher performance than their transformer counterparts: Sepformer, Conformer, and VALL-E. They are more efficient than transformers in memory and speed for speech longer than a threshold duration, inversely related to the resolution of a speech token. Mamba for separation is the most efficient, and Mamba for recognition is the least. Further, we show that Mamba is not more efficient than transformer for speech shorter than the threshold duration and performs worse in models that require joint modeling of text and speech, such as cross or masked attention of two inputs. Therefore, we argue that the superiority of Mamba or transformer depends on particular problems and models. Code available1

Blind Speech Separation (BSS) aims to separate multiple speech sources from audio mixtures recorded by a microphone array. The problem is challenging because it is a blind inverse problem, i.e., the microphone array geometry, the room impulse response (RIR), and the speech sources, are all unknown. We propose ArrayDPS to solve the BSS problem in an unsupervised, array-agnostic, and generative manner. The core idea builds on diffusion posterior sampling (DPS), but unlike DPS where the likelihood is tractable, ArrayDPS must approximate the likelihood by formulating a separate optimization problem. The solution to the optimization approximates room acoustics and the relative transfer functions between microphones. These approximations, along with the diffusion priors, iterate through the ArrayDPS sampling process and ultimately yield separated voice sources. We only need a simple single-speaker speech diffusion model as a prior, along with the mixtures recorded at the microphones; no microphone array information is necessary. Evaluation results show that ArrayDPS outperforms all baseline unsupervised methods while being comparable to supervised methods in terms of SDR. Audio demos and codes are provided at: https://arraydps.github.io/ArrayDPSDemo/ and https://github.com/ArrayDPS/ArrayDPS.

Yinghao Aaron Li, Xilin Jiang, Cong Han, Nima Mesgarani. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

In this study, we present a simple multi-channel framework for contrastive learning (MC-SimCLR) to encode 'what' and 'where' of spatial audios. MC-SimCLR learns joint spectral and spatial representations from unlabeled spatial audios, thereby enhancing both event classification and sound localization in downstream tasks. At its core, we propose a multi-level data augmentation pipeline that augments different levels of audio features, including waveforms, Mel spectrograms, and generalized cross-correlation (GCC) features. In addition, we introduce simple yet effective channel-wise augmentation methods to randomly swap the order of the microphones and mask Mel and GCC channels. By using these augmentations, we find that linear layers on top of the learned representation significantly outperform supervised models in terms of both event classification accuracy and localization error. We also perform a comprehensive analysis of the effect of each augmentation method and a comparison of the fine-tuning performance using different amounts of labeled data.

Transformers have revolutionized deep learning across various tasks, including audio representation learning, due to their powerful modeling capabilities. However, they often suffer from quadratic complexity in both GPU memory usage and computational inference time, affecting their efficiency. Recently, state space models (SSMs) like Mamba have emerged as a promising alternative, offering a more efficient approach by avoiding these complexities. Given these advantages, we explore the potential of SSM-based models in audio tasks. In this paper, we introduce Self-Supervised Audio Mamba (SSAMBA), the first self-supervised, attention-free, and SSM-based model for audio representation learning. SSAMBA leverages the bidirectional Mamba to capture complex audio patterns effectively. We incorporate a self-supervised pretraining framework that optimizes both discriminative and generative objectives, enabling the model to learn robust audio representations from large-scale, unlabeled datasets. We evaluated SSAMBA on various tasks such as audio classification, keyword spotting, speaker identification, and emotion recognition. Our results demonstrate that SSAMBA outperforms the Self-Supervised Audio Spectrogram Transformer (SSAST) in most tasks. Notably, SSAMBA is approximately 92.7% faster in batch inference speed and 95.4% more memory-efficient than SSAST for the tiny model size with an input token size of 22k. These efficiency gains, combined with superior performance, underscore the effectiveness of SSAMBA’s architectural innovation, making it a compelling choice for a wide range of audio processing applications. Code at https://github.com/SiavashShams/ssamba.

In recent years, we have observed a rapid advancement in speech language models (SpeechLLMs), catching up with humans’ listening and reasoning abilities. SpeechLLMs have demonstrated impressive spoken dialog question-answering (SQA) performance in benchmarks like Gaokao, the English listening test of the college entrance exam in China, which seemingly requires understanding both the spoken content and voice characteristics of speakers in a conversation. However, after carefully examining Gaokao’s questions, we find the correct answers to many questions can be inferred from the conversation transcript alone, i.e. without speaker segmentation and identification. Our evaluation of state-of-the-art models Qwen-Audio and WavLLM on both Gaokao and our proposed “What Do You Like?” dataset shows a significantly higher accuracy in these context-based questions than in identity-critical questions, which can only be answered reliably with correct speaker identification. The results and analysis suggest that when solving SQA, the current SpeechLLMs exhibit limited speaker awareness from the audio and behave similarly to an LLM reasoning from the conversation transcription without sound. We propose that tasks focused on identity-critical questions could offer a more accurate evaluation framework of SpeechLLMs in SQA.

Large-scale pre-trained language models have been shown to be helpful in improving the naturalness of text-to-speech (TTS) models by enabling them to produce more naturalistic prosodic patterns. However, these models are usually word-level or sup-phoneme-level and jointly trained with phonemes, making them inefficient for the downstream TTS task where only phonemes are needed. In this work, we propose a phoneme-level BERT (PL-BERT) with a pretext task of predicting the corresponding graphemes along with the regular masked phoneme predictions. Subjective evaluations show that our phoneme-level BERT encoder has significantly improved the mean opinion scores (MOS) of rated naturalness of synthesized speech compared with the state-of-the-art (SOTA) StyleTTS baseline on out-of-distribution (OOD) texts.

Lifelong audio feature extraction involves learning new sound classes incrementally, which is essential for adapting to new data distributions over time.However, optimizing the model only on new data can lead to catastrophic forgetting of previously learned tasks, which undermines the model's ability to perform well over the long term.This paper introduces a new approach to continual audio representation learning called DeCoR.Unlike other methods that store previous data, features, or models, DeCoR indirectly distills knowledge from an earlier model to the latest by predicting quantization indices from a delayed codebook.We demonstrate that DeCoR improves acoustic scene classification accuracy and integrates well with continual self-supervised representation learning.Our approach introduces minimal storage and computation overhead, making it a lightweight and efficient solution for continual learning.

In this article, we work on a sound recognition system that continually incorporates new sound classes. Our main goal is to develop a framework where the model can be updated without relying on labeled data. For this purpose, we propose adopting representation learning, where an encoder is trained using unlabeled data. This learning framework enables the study and implementation of a practically relevant use case where only a small amount of the labels is available in a continual learning context. We also make the empirical observation that a similarity-based representation learning method within this framework is robust to forgetting even if no explicit mechanism against forgetting is employed. We show that this approach obtains similar performance compared to several distillation-based continual learning methods when employed on self-supervised representation learning methods.