Lingwei Meng, Long Zhou, Shujie Liu, Sanyuan Chen, Bing Han, Shujie Hu, Yanqing Liu, Jinyu Li, Sheng Zhao, Xixin Wu, Helen M. Meng, Furu Wei. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Quantifying audio aesthetics is challenging due to its subjective nature, influenced by human perception and cultural context. Traditional methods rely on human listeners, leading to inconsistencies and high resource demands. This paper addresses the growing need for automated systems capable of predicting audio aesthetics without human intervention. Such systems are crucial for applications like data filtering, pseudo-labeling, and evaluating generative models.In this paper, we propose new annotation guidelines that break down human listening perspectives into four axes and develop no-reference, peritem prediction models for more nuanced audio quality assessment. Our models are evaluated against human mean opinion scores (MOS) and existing methods, demonstrating comparable or superior performance. This research not only advances the field of audio aesthetics but also provides open-source models and datasets to facilitate future work and benchmarking.

Built upon vector quantization (VQ), discrete audio codec models have achieved great success in audio compression and autoregressive audio generation. However, existing models face substantial challenges in perceptual quality and signal distortion, especially when operating in extremely low bandwidth, rooted in the sensitivity of the VQ codebook to noise. This degradation poses significant challenges for several downstream tasks, such as codec-based speech synthesis. To address this issue, we propose a novel VQ method, Normal Distribution-based Vector Quantization (NDVQ), by introducing an explicit margin between the VQ codes via learning a variance. Specifically, our approach involves mapping the waveform to a latent space and quantizing it by selecting the most likely normal distribution, with each codebook entry representing a unique normal distribution defined by its mean and variance. Using these distribution-based VQ codec codes, a decoder reconstructs the input waveform. NDVQ is trained with additional distribution-related losses, alongside reconstruction and discrimination losses. Experiments demonstrate that NDVQ outperforms existing audio compression baselines, such as EnCodec, in terms of audio quality and zero-shot TTS, particularly in very low bandwidth scenarios.

Recent advancements in generative speech models based on audio-text prompts have enabled remarkable innovations like high-quality zero-shot text-to-speech. However, existing models still face limitations in handling diverse audio-text speech generation tasks involving transforming input speech and processing audio captured in adverse acoustic conditions. This paper introduces SpeechX, a versatile speech generation model capable of zero-shot TTS and various speech transformation tasks, dealing with both clean and noisy signals. SpeechX combines neural codec language modeling with multi-task learning using task-dependent prompting, enabling unified and extensible modeling and providing a consistent way for leveraging textual input in speech enhancement and transformation tasks. Experimental results show SpeechX's efficacy in various tasks, including zero-shot TTS, noise suppression, target speaker extraction, speech removal, and speech editing with or without background noise, achieving comparable or superior performance to specialized models across tasks.

How to boost speech pre-training with textual data is an unsolved problem due to the fact that speech and text are very different modalities with distinct characteristics. In this paper, we propose a cross-modalSpeechandLanguageModel (SpeechLM) to explicitly align speech and text pre-training with a pre-defined unified discrete representation. Specifically, we introduce two alternative discrete tokenizers to bridge the speech and text modalities, including phoneme-unit and hidden-unit tokenizers, which can be trained using unpaired speech or a small amount of paired speech-text data. Based on the trained tokenizers, we convert the unlabeled speech and text data into tokens of phoneme units or hidden units. The pre-training objective is designed to unify the speech and the text into the same discrete semantic space with a unified Transformer network. We evaluate SpeechLM on various spoken language processing tasks including speech recognition, speech translation, and universal representation evaluation framework SUPERB, demonstrating significant improvements on content-related tasks. Code and models are available athttps://aka.ms/SpeechLM.

We introduce a self-supervised learning (SSL) framework BEATs for general audio representation pre-training, where we optimize an acoustic tokenizer and an audio SSL model by iterations. Unlike the previous audio SSL models that employ reconstruction loss for pre-training, our audio SSL model is trained with the discrete label prediction task, where the labels are generated by a semantic-rich acoustic tokenizer. We propose an iterative pipeline to jointly optimize the tokenizer and the pre-trained model, aiming to abstract high-level semantics and discard the redundant details for audio. The experimental results demonstrate our acoustic tokenizers can generate discrete labels with rich audio semantics and our audio SSL models achieve state-of-the-art (SOTA) results across various audio classification benchmarks, even outperforming previous models that use more training data and model parameters significantly. Specifically, we set a new SOTA mAP 50.6% on AudioSet-2M without using any external data, and 98.1% accuracy on ESC-50. The code and pre-trained models are available at https://aka.ms/beats.

In order to better understand the rationale behind model behavior, recent works have exploited providing interpretation to support the inference prediction.However, existing methods tend to provide human-unfriendly interpretation, and are prone to sub-optimal performance due to one-side promotion, i.e. either inference promotion with interpretation or vice versa.In this paper, we propose a multi-level Mutual Promotion mechanism for self-evolved Inference and sentence-level Interpretation (MPII).Specifically, from the model-level, we propose a Step-wise Integration Mechanism to jointly perform and deeply integrate inference and interpretation in an autoregressive manner.From the optimizationlevel, we propose an Adversarial Fidelity Regularization to improve the fidelity between inference and interpretation with the Adversarial Mutual Information training strategy.Extensive experiments on NLI and CQA tasks reveal that the proposed MPII approach can significantly outperform baseline models for both the inference performance and the interpretation quality. 1

The speech representations learned from large-scale unlabeled data have shown better generalizability than those from supervised learning and thus attract a lot of interest to be applied for various downstream tasks. In this paper, we explore the limits of speech representations learned by different self-supervised objectives and datasets for automatic speaker verification (ASV), especially with a well-recognized SOTA ASV model, ECAPA-TDNN [1], as a downstream model. The representations from all hidden layers of the pre-trained model are firstly averaged with learnable weights and then fed into the ECAPA-TDNN as input features. The experimental results on Voxceleb dataset show that the weighted average representation is significantly superior to FBank, a conventional handcrafted feature for ASV. Our best single system achieves 0.537%, 0.569%, and 1.180% equal error rate (EER) on the three official trials of VoxCeleb1, separately. Accordingly, the ensemble system with three pre-trained models can further improve the EER to 0.479%, 0.536% and 1.023%. Among the three evaluation trials, our best system outperforms the winner system [2] of the VoxCeleb Speaker Recognition Challenge 2021 (VoxSRC2021) on the VoxCeleb1-E trial.

Self-supervised learning (SSL) is a long-standing goal for speech processing, since it utilizes large-scale unlabeled data and avoids extensive human labeling. Recent years have witnessed great successes in applying self-supervised learning in speech recognition, while limited exploration was attempted in applying SSL for modeling speaker characteristics. In this paper, we aim to improve the existing SSL framework for speaker representation learning. Two methods are introduced for enhancing the unsupervised speaker information extraction. First, we apply multi-task learning to the current SSL framework, where we integrate utterance-wise contrastive loss with the SSL objective function. Second, for better speaker discrimination, we propose an utterance mixing strategy for data augmentation, where additional overlapped utterances are created unsupervisely and incorporated during training. We integrate the proposed methods into the HuBERT framework. Experiment results on the SUPERB benchmark show that the proposed system achieves state-of-the-art performance in universal representation learning, especially for speaker identification oriented tasks. An ablation study is performed verifying the efficacy of each proposed method. Finally, we scale up the training dataset to 94 thousand hours of public audio data and achieve further performance improvement in all SUPERB tasks.

Recently, pioneer work finds that self-supervised pre-training methods can improve multiple downstream speech tasks, because the model utilizes bottom layers to learn speaker-related information and top layers to encode content-related information. Since the network capacity is limited, we believe the speech recognition performance could be further improved if the model is dedicated to audio content information learning. To this end, we propose Intermediate Layer Supervision for Self-Supervised Learning (ILS-SSL), which forces the model to concentrate on content information as much as possible by adding an additional SSL loss on the intermediate layers. Experiments on LibriSpeech test-other set show that our method outperforms HuBERT significantly, which achieves a 23.5%/11.6% relative word error rate reduction in the w/o language model setting for Base/Large models. Detailed analysis shows the bottom layers of our model have a better correlation with phonetic units, which is consistent with our intuition and explains the success of our method for ASR. We will release our code and model at https://github.com/microsoft/UniSpeech.

Recently, self-supervised learning (SSL) has demonstrated strong performance in speaker recognition, even if the pretraining objective is designed for speech recognition.In this paper, we study which factor leads to the success of selfsupervised learning on speaker-related tasks, e.g.speaker verification (SV), through a series of carefully designed experiments.Our empirical results on the Voxceleb-1 dataset suggest that the benefit of SSL to SV task is from a combination of mask speech prediction loss, data scale, and model size, while the SSL quantizer has a minor impact.We further employ the integrated gradients attribution method and loss landscape visualization to understand the effectiveness of self-supervised learning for speaker recognition performance.

Recently, masked prediction pre-training has seen remarkable progress in self-supervised learning (SSL) for speech recognition.It usually requires a codebook obtained in an unsupervised way, making it less accurate and difficult to interpret.We propose two supervision-guided codebook generation approaches to improve automatic speech recognition (ASR) performance and also the pre-training efficiency, either through decoding with a hybrid ASR system to generate phoneme-level alignments (named PBERT), or performing clustering on the supervised speech features extracted from an end-to-end CTC model (named CTC clustering).Both the hybrid and CTC models are trained on the same small amount of labeled speech as used in fine-tuning.Experiments demonstrate significant superiority of our methods to various SSL and self-training baselines, with up to 17.0% relative WER reduction.Our pre-trained models also show good transferability in a non-ASR speech task.

There is a surge in interest in self-supervised learning approaches for end-to-end speech encoding in recent years as they have achieved great success. Especially, WavLM showed state-of-the-art performance on various speech processing tasks. To better understand the efficacy of self-supervised learning models for speech enhancement, in this work, we design and conduct a series of experiments with three resource conditions by combining WavLM and two high-quality speech enhancement systems. Also, We propose a regression-based WavLM training objective and a noise-mixing data configuration to further boost the downstream enhancement performance. The experiments on the DNS challenge dataset and a simulation dataset show that the WavLM benefits the speech enhancement task in terms of both speech quality and speech recognition accuracy, especially for low fine-tuning resources. For the high fine-tuning resource condition, only the word error rate is substantially improved.

Slot filling, a fundamental module of spoken language understanding, often suffers from insufficient quantity and diversity of training data. To remedy this, we propose a novel Cluster-to-Cluster generation framework for Data Augmentation (DA), named C2C-GenDA. It enlarges the training set by reconstructing existing utterances into alternative expressions while keeping semantic. Different from previous DA works that reconstruct utterances one by one independently, C2C-GenDA jointly encodes multiple existing utterances of the same semantics and simultaneously decodes multiple unseen expressions. Jointly generating multiple new utterances allows to consider the relations between generated instances and encourages diversity. Besides, encoding multiple existing utterances endows C2C with a wider view of existing expressions, helping to reduce generation that duplicates existing data. Experiments on ATIS and Snips datasets show that instances augmented by C2C-GenDA improve slot filling by 7.99 (11.9%↑) and 5.76 (13.6%↑) F-scores respectively, when there are only hundreds of training utterances. Code: https://github.com/Sanyuan-Chen/C2C-DA.

With its strong modeling capacity that comes from a multi-head and multi-layer structure, Transformer is a very powerful model for learning a sequential representation and has been successfully applied to speech separation recently. However, multi-channel speech separation sometimes does not necessarily need such a heavy structure for all time frames especially when the cross-talker challenge happens only occasionally. For example, in conversation scenarios, most regions contain only a single active speaker, where the separation task downgrades to a single speaker enhancement problem. It turns out that using a very deep network structure for dealing with signals with a low overlap ratio not only negatively affects the inference efficiency but also hurts the separation performance. To deal with this problem, we propose an early exit mechanism, which enables the Transformer model to handle different cases with adaptive depth. Experimental results indicate that not only does the early exit mechanism accelerate the inference, but it also improves the accuracy.

Continuous speech separation was recently proposed to deal with the overlapped speech in natural conversations. While it was shown to significantly improve the speech recognition performance for multichannel conversation transcription, its effectiveness has yet to be proven for a single-channel recording scenario. This paper examines the use of Conformer architecture in lieu of recurrent neural networks for the separation model. Conformer allows the separation model to efficiently capture both local and global context information, which is helpful for speech separation. Experimental results using the LibriCSS dataset show that the Conformer separation model achieves the state of the art results for both single-channel and multi-channel settings. Results for real meeting recordings are also presented, showing significant performance gains in both word error rate (WER) and speaker-attributed WER.

This paper describes the Microsoft speaker diarization system for monaural multi-talker recordings in the wild, evaluated at the diarization track of the VoxCeleb Speaker Recognition Challenge (VoxSRC) 2020. We will first explain our system design to address issues in handling real multi-talker recordings. We then present the details of the components, which include Res2Net-based speaker embedding extractor, conformer-based continuous speech separation with leakage filtering, and a modified DOVER (short for Diarization Output Voting Error Reduction) method for system fusion. We evaluate the systems with the data set provided by VoxSRC challenge 2020, which contains real-life multi-talker audio collected from YouTube. Our best system achieves 3.71% and 6.23% of the diarization error rate (DER) on development set and evaluation set, respectively, being ranked the 1st at the diarization track of the challenge.

Transformer has been successfully applied to speech separation recently with its strong long-dependency modeling capacity using a self-attention mechanism. However, Transformer tends to have heavy run-time costs due to the deep encoder layers, which hinders its deployment on edge devices. A small Transformer model with fewer encoder layers is preferred for computational efficiency, but it is prone to performance degradation. In this paper, an ultra fast speech separation Transformer model is proposed to achieve both better performance and efficiency with teacher student learning (T-S learning). We introduce layer-wise T-S learning and objective shifting mechanisms to guide the small student model to learn intermediate representations from the large teacher model. Compared with the small Transformer model trained from scratch, the proposed T-S learning method reduces the word error rate (WER) by more than 5% for both multi-channel and single-channel speech separation on LibriCSS dataset. Utilizing more unlabeled speech data, our ultra fast speech separation models achieve more than 10% relative WER reduction.

Speech separation has been successfully applied as a frontend processing module of conversation transcription systems thanks to its ability to handle overlapped speech and its flexibility to combine with downstream tasks such as automatic speech recognition (ASR). However, a speech separation model often introduces target speech distortion, resulting in a sub-optimum word error rate (WER). In this paper, we describe our efforts to improve the performance of a single channel speech separation system. Specifically, we investigate a two-stage training scheme that firstly applies a feature level optimization criterion for pretraining, followed by an ASR-oriented optimization criterion using an end-to-end (E2E) speech recognition model. Meanwhile, to keep the model light-weight, we introduce a modified teacher-student learning technique for model compression. By combining those approaches, we achieve a absolute average WER improvement of 2.70% and 0.77% using models with less than 10M parameters compared with the previous state-of-the-art results on the LibriCSS dataset for utterance-wise evaluation and continuous evaluation, respectively

Deep pretrained language models have achieved great success in the way of pretraining first and then fine-tuning.But such a sequential transfer learning paradigm often confronts the catastrophic forgetting problem and leads to sub-optimal performance.To fine-tune with less forgetting, we propose a recall and learn mechanism, which adopts the idea of multi-task learning and jointly learns pretraining tasks and downstream tasks.Specifically, we propose a Pretraining Simulation mechanism to recall the knowledge from pretraining tasks without data, and an Objective Shifting mechanism to focus the learning on downstream tasks gradually.Experiments show that our method achieves state-of-the-art performance on the GLUE benchmark.Our method also enables BERT-base to achieve better performance than directly fine-tuning of BERT-large.Further, we provide the open-source RECADAM optimizer, which integrates the proposed mechanisms into Adam optimizer, to facility the NLP community.