Soft Clustering Anchors for Self-Supervised Speech Representation Learning in Joint Embedding Prediction Architectures
Carnegie Mellon University（卡内基梅隆大学）；New York University（纽约大学）；James Silberrad Brown Center for AI（詹姆斯·西伯拉德·布朗人工智能中心）；Columbia University（哥伦比亚大学）；Northeastern University（东北大学）；Stanford University（斯坦福大学）；Amazon GenAI（亚马逊生成人工智能）
Georgios Ioannides, Adrian Kieback, Judah Goldfeder, Linsey Pang, Aman Chadha, Aaron Elkins, Yann LeCun, Ravid Shwartz-Ziv
联合嵌入预测架构（JEPA）提供了一种有前途的自监督语音表示学习方法，但缺乏显式锚定会导致表示崩溃。我们提出GMM-Anchored JEPA，该方法在对数梅尔谱图上拟合一个高斯混合模型，并在训练过程中使用其冻结的软后验作为辅助目标。衰减的监督计划使GMM正则化在早期训练中占主导地位，然后逐渐让位于JEPA目标。与HuBERT和WavLM不同，我们的方法使用软而非硬分配对输入特征进行一次聚类。在约5万小时的语音数据上，GMM锚定将ASR（28.68% vs. 33.22% WER）、情感识别（67.76% vs. 65.46%）和槽填充（64.7% vs. 59.1% F1）相比WavLM风格基线（计算匹配）有所提升。聚类分析显示，GMM锚定表示的熵达到98%，而WavLM风格为31%，表明聚类利用更加均匀。代码可在https://github.com/gioannides/clustering-anchored-jepa上获得。
英文摘要：
Joint Embedding Predictive Architectures (JEPA) offer a promising approach to self-supervised speech representation learning, but suffer from representation collapse without explicit grounding. We propose GMM-Anchored JEPA, which fits a Gaussian Mixture Model once on log-mel spectrograms and uses its frozen soft posteriors as auxiliary targets throughout training. A decaying supervision schedule allows GMM regularization to dominate early training before gradually yielding to the JEPA objective. Unlike HuBERT and WavLM, which require iterative re-clustering, our approach clusters input features once with soft rather than hard assignments. On ~50k hours of speech, GMM anchoring improves ASR (28.68% vs. 33.22% WER), emotion recognition (67.76% vs. 65.46%), and slot filling (64.7% vs. 59.1% F1) compared to a WavLM-style baseline with matched compute. Cluster analysis shows GMM-anchored representations achieve up to 98% entropy compared to 31% for WavLM-style, indicating substantially more uniform cluster utilization. Code is made available at https://github.com/gioannides/clustering-anchored-jepa.
Neural transducers have achieved human level performance on standard speech recognition benchmarks. However, their performance significantly degrades in the presence of cross-talk, especially when the primary speaker has a low signal-to-noise ratio. Anchored speech recognition refers to a class of methods that use information from an anchor segment (e.g., wake-words) to recognize device-directed speech while ignoring interfering background speech. In this paper, we investigate anchored speech recognition to make neural transducers robust to background speech. We extract context information fro…
Voice-controlled house-hold devices, like Amazon Echo or Google Home, face the problem of performing speech recognition of device-directed speech in the presence of interfering background speech, i.e., background noise and interfering speech from another person or media device in proximity need to be ignored. We propose two end-to-end models to tackle this problem with information extracted from the "anchored segment". The anchored segment refers to the wake-up word part of an audio stream, which contains valuable speaker information that can be used to suppress interfering speech and backgrou…
Microsoft Research Asia（微软亚洲研究院）；University of Technology Sydney（悉尼科技大学）
Continuous Speech Keyword Spotting (CSKWS) is a task to detect predefined keywords in a continuous speech. In this paper, we regard CSKWS as a one-dimensional object detection task and propose a novel anchor-free detector, named AF-KWS, to solve the problem. AF-KWS directly regresses the center locations and lengths of the keywords through a single-stage deep neural network. In particular, AF-KWS is tailored for this speech task as we introduce an auxiliary unknown class to exclude other words from non-speech or silent background. We have built two benchmark datasets named LibriTop-20 and cont…
Advanced Clustering Techniques for Speech Signal Enhancement: A Review and Metanalysis of Fuzzy C-Means, K-Means, and Kernel Fuzzy C-Means Methods
Artificial Intelligence and Innovation Centre, University of Kurdistan Hewler（库尔德斯坦赫勒大学人工智能与创新中心）；Computer Science and Engineering Department, University of Kurdistan Hewler（库尔德斯坦赫勒大学计算机科学与工程系）
Keyword Spotting (KWS) remains challenging to achieve the trade-off between small footprint and high accuracy. Recently proposed metric learning approaches improved the generalizability of models for the KWS task, and 1D-CNN based KWS models have achieved the state-of-the-arts (SOTA) in terms of model size. However, for metric learning, due to data limitations, the speech anchor is highly susceptible to the acoustic environment and speakers. Also, we note that the 1D-CNN models have limited capability to capture long-term temporal acoustic features. To address the above problems, we propose to…
Normal-Anchored First-Order Model-Agnostic Meta-Learning based Whisper Fine-Tuning for Enhancing Fairness of Cleft Lip and Palate Speech Recognition
Indian Institute of Technology Guwahati（印度古瓦哈蒂印度理工学院）；University of Eastern Finland（东芬兰大学）；Brigham and Women's Hospital, Harvard Medical School（哈佛医学院布莱根妇女医院）；Indian Institute of Information Technology Dharwad (IIIT Dharwad)（达尔瓦德印度信息技术学院（IIIT达尔瓦德））
Discrete Speech Unit Extraction via Independent Component Analysis
National Institute of Advanced Industrial Science and Technology（日本国立先进工业科学技术研究院）；Carnegie Mellon University（卡内基梅隆大学）；Toyohashi University of Technology（丰桥技术科学大学）
Significant performance degradation of automatic speech recognition (ASR) systems is observed when the audio signal contains cross-talk. One of the recently proposed approaches to solve the problem of multi-speaker ASR is the deep clustering (DPCL) approach. Combining DPCL with a state-of-the-art hybrid acoustic model, we obtain a word error rate (WER) of 16.5 % on the commonly used wsj0-2mix dataset, which is the best performance reported thus far to the best of our knowledge. The wsj0-2mix dataset contains simulated cross-talk where the speech of multiple speakers overlaps for almost the ent…
The Graduate Center, City University of New York（纽约城市大学研究生中心）；Brooklyn College, City University of New York（纽约城市大学布鲁克林学院）
Recent works have shown that Deep Recurrent Neural Networks using the LSTM architecture can achieve strong single-channel speech enhancement by estimating time-frequency masks. However, these models do not naturally generalize to multi-channel inputs from varying microphone configurations. In contrast, spatial clustering techniques can achieve such generalization but lack a strong signal model. Our work proposes a combination of the two approaches. By using LSTMs to enhance spatial clustering based time-frequency masks, we achieve both the signal modeling performance of multiple single-channel…
S-JEPA : Soft Clustering Anchors for Self-Supervised Speech Representation Learning
Carnegie Mellon University（卡内基梅隆大学）；New York University（纽约大学）；James Silberrad Brown Center for AI（詹姆斯·西尔伯拉德·布朗人工智能中心）；Columbia University（哥伦比亚大学）；Northeastern University（东北大学）；Stanford University（斯坦福大学）；Amazon GenAI（亚马逊生成式人工智能）
In recent studies, self-supervised pre-trained models tend to outperform supervised pre-trained models in transfer learning. In particular, self-supervised learning (SSL) of utterance-level speech representation can be used in speech applications that require discriminative representation of consistent attributes within an utterance: speaker, language, emotion, and age. Existing frame-level self-supervised speech representation, e.g., wav2vec, can be used as utterance-level representation with pooling, but the models are usually large. There are also SSL techniques to learn utterance-level rep…
Considering the abundance of unlabeled speech data and the high labeling costs, unsupervised learning methods can be essential for better system development. One of the most successful methods is contrastive self-supervised methods, which require negative sampling: sampling alternative samples to contrast with the current sample (anchor). However, it is hard to ensure if all the negative samples belong to classes different from the anchor class without labels. This paper applies a non-contrastive self-supervised learning method on an unlabeled speech corpus to learn utterance-level embeddings.…
Self-Powered LLM Modality Expansion for Large Speech-Text Models
Institute of Computing and Intelligence, Harbin Institute of Technology（哈尔滨工业大学计算与智能学院）；Faculty of Computing, Harbin Institute of Technology（哈尔滨工业大学计算机学院）；The University of Sydney（悉尼大学）；Nanyang Technological University（南洋理工大学）
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；CAS Center for Excellence in Brain Science and Intelligence Technology（中国科学院脑科学与智能技术卓越创新中心）
Monaural speech dereverberation is a very challenging task because no spatial cues can be used. When the additive noises exist, this task becomes more challenging. In this paper, we propose a joint training method for simultaneous speech denoising and dereverberation using deep embedding features, which is based on the deep clustering (DC). DC is a state-of-the-art method for speech separation that includes embedding learning and K-means clustering. As for our proposed method, it contains two stages: denoising and dereverberation. At the denoising stage, the DC network is leveraged to extract …
Self-supervised approaches for speech representation learning are challenged by three unique problems: (1) there are multiple sound units in each input utterance, (2) there is no lexicon of input sound units during the pre-training phase, and (3) sound units have variable lengths with no explicit segmentation. To deal with these three problems, we propose the Hidden-Unit BERT (HuBERT) approach for self-supervised speech representation learning, which utilizes an offline clustering step to provide aligned target labels for a BERT-like prediction loss. A key ingredient of our approach is applyin…
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 mo…
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；CAS Center for Excellence in Brain Science and Intelligence Technology（中国科学院脑科学与智能技术卓越创新中心）；University of Chinese Academy of Sciences（中国科学院大学）
Deep clustering (DC) and utterance-level permutation invariant training (uPIT) have been demonstrated promising for speaker-independent speech separation. DC is usually formulated as two-step processes: embedding learning and embedding clustering, which results in complex separation pipelines and a huge obstacle in directly optimizing the actual separation objectives. As for uPIT, it only minimizes the chosen permutation with the lowest mean square error, doesn't discriminate it with other permutations. In this paper, we propose a discriminative learning method for speaker-independent speech s…
DAS-PMVC: A Framework for Partial Multi-View Clustering via Dual Alignment and Structure Enhancement
Dalian University of Technology（大连理工大学）；Inspur Group Co., Ltd.（浪潮集团有限公司）；China University of Mining and Technology（中国矿业大学）；The University of Warwick（华威大学）
VAPO: End-to-end Slide-Enhanced Speech Recognition with Omni-modal Large Language Models
Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence（北京交通数据挖掘与具身智能重点实验室）；Unisound AI Technology Co., Ltd.（Unisound人工智能技术有限公司）；State Key Laboratory of AI Safety, Beijing（人工智能安全国家重点实验室）
This paper describes system setup of our submission to speaker diarisation track (Track 4) of VoxCeleb Speaker Recognition Challenge 2020. Our diarisation system consists of a well-trained neural network based speech enhancement model as pre-processing front-end of input speech signals. We replace conventional energy-based voice activity detection (VAD) with a neural network based VAD. The neural network based VAD provides more accurate annotation of speech segments containing only background music, noise, and other interference, which is crucial to diarisation performance. We apply agglomerat…
Deep clustering is a recently introduced deep learning architecture that uses discriminatively trained embeddings as the basis for clustering. It was recently applied to spectrogram segmentation, resulting in impressive results on speaker-independent multi-speaker separation. In this paper we extend the baseline system with an end-to-end signal approximation objective that greatly improves performance on a challenging speech separation. We first significantly improve upon the baseline system performance by incorporating better regularization, larger temporal context, and a deeper architecture,…
In recent years, deep-learning-based speech emotion recognition models have outperformed classical machine learning models. Previously, neural network designs, such as Multitask Learning, have accounted for variations in emotional expressions due to demographic and contextual factors. However, existing models face a few constraints: 1) they rely on a clear definition of domains (e.g. gender, noise condition, etc.) and the availability of domain labels; 2) they often attempt to learn domain-invariant features while emotion expressions can be domain-specific. In the present study, we propose the…
Facebook AI Research（脸书人工智能研究院）；University of Tübingen（图宾根大学）
We propose vq-wav2vec to learn discrete representations of audio segments through a wav2vec-style self-supervised context prediction task. The algorithm uses either a gumbel softmax or online k-means clustering to quantize the dense representations. Discretization enables the direct application of algorithms from the NLP community which require discrete inputs. Experiments show that BERT pre-training achieves a new state of the art on TIMIT phoneme classification and WSJ speech recognition.
In settings where only unlabelled speech data is available, zero-resource speech technology needs to be developed without transcriptions, pronunciation dictionaries, or language modelling text. There are two central problems in zero-resource speech processing: (i) finding frame-level feature representations which make it easier to discriminate between linguistic units (phones or words), and (ii) segmenting and clustering unlabelled speech into meaningful units. In this thesis, we argue that a combination of top-down and bottom-up modelling is advantageous in tackling these two problems. To add…
IBM Research AI（IBM研究院人工智能部门）；IBM T. J. Watson Research Center（IBM托马斯·J·沃森研究中心）
Evolutionary stochastic gradient descent (ESGD) was proposed as a population-based approach that combines the merits of gradient-aware and gradient-free optimization algorithms for superior overall optimization performance. In this paper we investigate a variant of ESGD for optimization of acoustic models for automatic speech recognition (ASR). In this variant, we assume the existence of a well-trained acoustic model and use it as an anchor in the parent population whose good "gene" will propagate in the evolution to the offsprings. We propose an ESGD algorithm leveraging the anchor models suc…
The Graduate Center, City University of New York（纽约市立大学研究生中心）；Brooklyn College, City University of New York（纽约市立大学布鲁克林学院）
Spatial clustering techniques can achieve significant multi-channel noise reduction across relatively arbitrary microphone configurations, but have difficulty incorporating a detailed speech/noise model. In contrast, LSTM neural networks have successfully been trained to recognize speech from noise on single-channel inputs, but have difficulty taking full advantage of the information in multi-channel recordings. This paper integrates these two approaches, training LSTM speech models to clean the masks generated by the Model-based EM Source Separation and Localization (MESSL) spatial clustering…
The Graduate Center, City University of New York（纽约城市大学研究生中心）；Brooklyn College, City University of New York（纽约城市大学布鲁克林学院）
Recurrent neural networks using the LSTM architecture can achieve significant single-channel noise reduction. It is not obvious, however, how to apply them to multi-channel inputs in a way that can generalize to new microphone configurations. In contrast, spatial clustering techniques can achieve such generalization, but lack a strong signal model. This paper combines the two approaches to attain both the spatial separation performance and generality of multichannel spatial clustering and the signal modeling performance of multiple parallel single-channel LSTM speech enhancers. The system is c…
Northwestern Polytechnical University（西北工业大学）；Bytedance AI Lab（字节跳动人工智能实验室）
Attention-based sequence-to-sequence (seq2seq) speech synthesis has achieved extraordinary performance. But a studio-quality corpus with manual transcription is necessary to train such seq2seq systems. In this paper, we propose an approach to build high-quality and stable seq2seq based speech synthesis system using challenging found data, where training speech contains noisy interferences (acoustic noise) and texts are imperfect speech recognition transcripts (textual noise). To deal with text-side noise, we propose a VQVAE based heuristic method to compensate erroneous linguistic feature with…
Dynamic time warping (DTW) can be used to compute the similarity between two sequences of generally differing length. We propose a modification to DTW that performs individual and independent pairwise alignment of feature trajectories. The modified technique, termed feature trajectory dynamic time warping (FTDTW), is applied as a similarity measure in the agglomerative hierarchical clustering of speech segments. Experiments using MFCC and PLP parametrisations extracted from TIMIT and from the Spoken Arabic Digit Dataset (SADD) show consistent and statistically significant improvements in the q…
Faculty of Engineering, Bar-Ilan University（巴伊兰大学工程学院）
In this study we present a mixture of deep experts (MoDE) neural-network architecture for single microphone speech enhancement. Our architecture comprises a set of deep neural networks (DNNs), each of which is an 'expert' in a different speech spectral pattern such as phoneme. A gating DNN is responsible for the latent variables which are the weights assigned to each expert's output given a speech segment. The experts estimate a mask from the noisy input and the final mask is then obtained as a weighted average of the experts' estimates, with the weights determined by the gating DNN. A soft sp…
Sentence level pronunciation assessment is important for Computer Assisted Language Learning (CALL). Traditional speech pronunciation assessment, based on the Goodness of Pronunciation (GOP) algorithm, has some weakness in assessing a speech utterance: 1) Phoneme GOP scores cannot be easily translated into a sentence score with a simple average for effective assessment; 2) The rank ordering information has not been well exploited in GOP scoring for delivering a robust assessment and correlate well with a human rater's evaluations. In this paper, we propose two new statistical features, average…
In certain applications such as zero-resource speech processing or very-low resource speech-language systems, it might not be feasible to collect speech activity detection (SAD) annotations. However, the state-of-the-art supervised SAD techniques based on neural networks or other machine learning methods require annotated training data matched to the target domain. This paper establish a clustering approach for fully unsupervised SAD useful for cases where SAD annotations are not available. The proposed approach leverages Hartigan dip test in a recursive strategy for segmenting the feature spa…
Speech separation has been very successful with deep learning techniques. Substantial effort has been reported based on approaches over spectrogram, which is well known as the standard time-and-frequency cross-domain representation for speech signals. It is highly correlated to the phonetic structure of speech, or "how the speech sounds" when perceived by human, but primarily frequency domain features carrying temporal behaviour. Very impressive work achieving speech separation over time domain was reported recently, probably because waveforms in time domain may describe the different realizat…
Recently, we proposed a novel speaker diarization method called End-to-End-Neural-Diarization-vector clustering (EEND-vector clustering) that integrates clustering-based and end-to-end neural network-based diarization approaches into one framework. The proposed method combines advantages of both frameworks, i.e. high diarization performance and handling of overlapped speech based on EEND, and robust handling of long recordings with an arbitrary number of speakers based on clustering-based approaches. However, the method was only evaluated so far on simulated 2-speaker meeting-like data. This p…
Deep clustering is a deep neural network-based speech separation algorithm that first trains the mixed component of signals with high-dimensional embeddings, and then uses a clustering algorithm to separate each mixture of sources. In this paper, we extend the baseline criterion of deep clustering with an additional regularization term to further improve the overall performance. This term plays a role in assigning a condition to the embeddings such that it gives less correlation to each embedding dimension, leading to better decomposition of the spectral bins. The regularization term helps to …
This paper presents a method for controlling the prosody at the phoneme level in an autoregressive attention-based text-to-speech system. Instead of learning latent prosodic features with a variational framework as is commonly done, we directly extract phoneme-level F0 and duration features from the speech data in the training set. Each prosodic feature is discretized using unsupervised clustering in order to produce a sequence of prosodic labels for each utterance. This sequence is used in parallel to the phoneme sequence in order to condition the decoder with the utilization of a prosodic en…
Speaker-aware source separation methods are promising workarounds for major difficulties such as arbitrary source permutation and unknown number of sources. However, it remains challenging to achieve satisfying performance provided a very short available target speaker utterance (anchor). Here we present a novel "deep extractor network" which creates an extractor point for the target speaker in a canonical high dimensional embedding space, and pulls together the time-frequency bins corresponding to the target speaker. The proposed model is different from prior works in that the canonical embed…
