Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

"speech": models, code, and papers

Normalizing Flow based Hidden Markov Models for Classification of Speech Phones with Explainability

Jul 01, 2021
Anubhab Ghosh, Antoine Honoré, Dong Liu, Gustav Eje Henter, Saikat Chatterjee

In pursuit of explainability, we develop generative models for sequential data. The proposed models provide state-of-the-art classification results and robust performance for speech phone classification. We combine modern neural networks (normalizing flows) and traditional generative models (hidden Markov models - HMMs). Normalizing flow-based mixture models (NMMs) are used to model the conditional probability distribution given the hidden state in the HMMs. Model parameters are learned through judicious combinations of time-tested Bayesian learning methods and contemporary neural network learning methods. We mainly combine expectation-maximization (EM) and mini-batch gradient descent. The proposed generative models can compute likelihood of a data and hence directly suitable for maximum-likelihood (ML) classification approach. Due to structural flexibility of HMMs, we can use different normalizing flow models. This leads to different types of HMMs providing diversity in data modeling capacity. The diversity provides an opportunity for easy decision fusion from different models. For a standard speech phone classification setup involving 39 phones (classes) and the TIMIT dataset, we show that the use of standard features called mel-frequency-cepstral-coeffcients (MFCCs), the proposed generative models, and the decision fusion together can achieve $86.6\%$ accuracy by generative training only. This result is close to state-of-the-art results, for examples, $86.2\%$ accuracy of PyTorch-Kaldi toolkit [1], and $85.1\%$ accuracy using light gated recurrent units [2]. We do not use any discriminative learning approach and related sophisticated features in this article.

* 12 pages, 4 figures 

  Access Paper or Ask Questions

Embedding Recurrent Layers with Dual-Path Strategy in a Variant of Convolutional Network for Speaker-Independent Speech Separation

Mar 25, 2022
Xue Yang, Changchun Bao

Speaker-independent speech separation has achieved remarkable performance in recent years with the development of deep neural network (DNN). Various network architectures, from traditional convolutional neural network (CNN) and recurrent neural network (RNN) to advanced transformer, have been designed sophistically to improve separation performance. However, the state-of-the-art models usually suffer from several flaws related to the computation, such as large model size, huge memory consumption and computational complexity. To find the balance between the performance and computational efficiency and to further explore the modeling ability of traditional network structure, we combine RNN and a newly proposed variant of convolutional network to cope with speech separation problem. By embedding two RNNs into basic block of this variant with the help of dual-path strategy, the proposed network can effectively learn the local information and global dependency. Besides, a four-staged structure enables the separation procedure to be performed gradually at finer and finer scales as the feature dimension increases. The experimental results on various datasets have proven the effectiveness of the proposed method and shown that a trade-off between the separation performance and computational efficiency is well achieved.

* Submitted to Interspeech 2022 

  Access Paper or Ask Questions

Detecting Online Hate Speech: Approaches Using Weak Supervision and Network Embedding Models

Jul 24, 2020
Michael Ridenhour, Arunkumar Bagavathi, Elaheh Raisi, Siddharth Krishnan

The ubiquity of social media has transformed online interactions among individuals. Despite positive effects, it has also allowed anti-social elements to unite in alternative social media environments (eg. like never before. Detecting such hateful speech using automated techniques can allow social media platforms to moderate their content and prevent nefarious activities like hate speech propagation. In this work, we propose a weak supervision deep learning model that - (i) quantitatively uncover hateful users and (ii) present a novel qualitative analysis to uncover indirect hateful conversations. This model scores content on the interaction level, rather than the post or user level, and allows for characterization of users who most frequently participate in hateful conversations. We evaluate our model on 19.2M posts and show that our weak supervision model outperforms the baseline models in identifying indirect hateful interactions. We also analyze a multilayer network, constructed from two types of user interactions in Gab(quote and reply) and interaction scores from the weak supervision model as edge weights, to predict hateful users. We utilize the multilayer network embedding methods to generate features for the prediction task and we show that considering user context from multiple networks help achieving better predictions of hateful users in Gab. We receive up to 7% performance gain compared to single layer or homogeneous network embedding models.

  Access Paper or Ask Questions

A Scalable Model Specialization Framework for Training and Inference using Submodels and its Application to Speech Model Personalization

Mar 23, 2022
Fadi Biadsy, Youzheng Chen, Xia Zhang, Oleg Rybakov, Andrew Rosenberg, Pedro J. Moreno

Model fine-tuning and adaptation have become a common approach for model specialization for downstream tasks or domains. Fine-tuning the entire model or a subset of the parameters using light-weight adaptation has shown considerable success across different specialization tasks. Fine-tuning a model for a large number of domains typically requires starting a new training job for every domain posing scaling limitations. Once these models are trained, deploying them also poses significant scalability challenges for inference for real-time applications. In this paper, building upon prior light-weight adaptation techniques, we propose a modular framework that enables us to substantially improve scalability for model training and inference. We introduce Submodels that can be quickly and dynamically loaded for on-the-fly inference. We also propose multiple approaches for training those Submodels in parallel using an embedding space in the same training job. We test our framework on an extreme use-case which is speech model personalization for atypical speech, requiring a Submodel for each user. We obtain 128x Submodel throughput with a fixed computation budget without a loss of accuracy. We also show that learning a speaker-embedding space can scale further and reduce the amount of personalization training data required per speaker.

* Submitted to INTERSPEECH 

  Access Paper or Ask Questions

ConcealNet: An End-to-end Neural Network for Packet Loss Concealment in Deep Speech Emotion Recognition

May 15, 2020
Mostafa M. Mohamed, Björn W. Schuller

Packet loss is a common problem in data transmission, including speech data transmission. This may affect a wide range of applications that stream audio data, like streaming applications or speech emotion recognition (SER). Packet Loss Concealment (PLC) is any technique of facing packet loss. Simple PLC baselines are 0-substitution or linear interpolation. In this paper, we present a concealment wrapper, which can be used with stacked recurrent neural cells. The concealment cell can provide a recurrent neural network (ConcealNet), that performs real-time step-wise end-to-end PLC at inference time. Additionally, extending this with an end-to-end emotion prediction neural network provides a network that performs SER from audio with lost frames, end-to-end. The proposed model is compared against the fore-mentioned baselines. Additionally, a bidirectional variant with better performance is utilised. For evaluation, we chose the public RECOLA dataset given its long audio tracks with continuous emotion labels. ConcealNet is evaluated on the reconstruction of the audio and the quality of corresponding emotions predicted after that. The proposed ConcealNet model has shown considerable improvement, for both audio reconstruction and the corresponding emotion prediction, in environments that do not have losses with long duration, even when the losses occur frequently.

* Submission for INTERSPEECH 2020. 4 Pages + 1 references page. 4 Tables, 3 Figures 

  Access Paper or Ask Questions

AraCOVID19-MFH: Arabic COVID-19 Multi-label Fake News and Hate Speech Detection Dataset

May 07, 2021
Mohamed Seghir Hadj Ameur, Hassina Aliane

Along with the COVID-19 pandemic, an "infodemic" of false and misleading information has emerged and has complicated the COVID-19 response efforts. Social networking sites such as Facebook and Twitter have contributed largely to the spread of rumors, conspiracy theories, hate, xenophobia, racism, and prejudice. To combat the spread of fake news, researchers around the world have and are still making considerable efforts to build and share COVID-19 related research articles, models, and datasets. This paper releases "AraCOVID19-MFH" a manually annotated multi-label Arabic COVID-19 fake news and hate speech detection dataset. Our dataset contains 10,828 Arabic tweets annotated with 10 different labels. The labels have been designed to consider some aspects relevant to the fact-checking task, such as the tweet's check worthiness, positivity/negativity, and factuality. To confirm our annotated dataset's practical utility, we used it to train and evaluate several classification models and reported the obtained results. Though the dataset is mainly designed for fake news detection, it can also be used for hate speech detection, opinion/news classification, dialect identification, and many other tasks.

  Access Paper or Ask Questions

User-Level Differential Privacy against Attribute Inference Attack of Speech Emotion Recognition in Federated Learning

Apr 05, 2022
Tiantian Feng, Raghuveer Peri, Shrikanth Narayanan

Many existing privacy-enhanced speech emotion recognition (SER) frameworks focus on perturbing the original speech data through adversarial training within a centralized machine learning setup. However, this privacy protection scheme can fail since the adversary can still access the perturbed data. In recent years, distributed learning algorithms, especially federated learning (FL), have gained popularity to protect privacy in machine learning applications. While FL provides good intuition to safeguard privacy by keeping the data on local devices, prior work has shown that privacy attacks, such as attribute inference attacks, are achievable for SER systems trained using FL. In this work, we propose to evaluate the user-level differential privacy (UDP) in mitigating the privacy leaks of the SER system in FL. UDP provides theoretical privacy guarantees with privacy parameters $\epsilon$ and $\delta$. Our results show that the UDP can effectively decrease attribute information leakage while keeping the utility of the SER system with the adversary accessing one model update. However, the efficacy of the UDP suffers when the FL system leaks more model updates to the adversary. We make the code publicly available to reproduce the results in

  Access Paper or Ask Questions

A Machine Learning Approach to Detect Suicidal Ideation in US Veterans Based on Acoustic and Linguistic Features of Speech

Sep 27, 2020
Vaibhav Sourirajan, Anas Belouali, Mary Ann Dutton, Matthew Reinhard, Jyotishman Pathak

Preventing Veteran suicide is a national priority. The US Department of Veterans Affairs (VA) collects, analyzes, and publishes data to inform suicide prevention strategies. Current approaches for detecting suicidal ideation mostly rely on patient self report which are inadequate and time consuming. In this research study, our goal was to automate suicidal ideation detection from acoustic and linguistic features of an individual's speech using machine learning (ML) algorithms. Using voice data collected from Veterans enrolled in a large interventional study on Gulf War Illness at the Washington DC VA Medical Center, we conducted an evaluation of the performance of different ML approaches in achieving our objective. By fitting both classical ML and deep learning models to the dataset, we identified the algorithms that were most effective for each feature set. Among classical machine learning algorithms, the Support Vector Machine (SVM) trained on acoustic features performed best in classifying suicidal Veterans. Among deep learning methods, the Convolutional Neural Network (CNN) trained on the linguistic features performed best. Our study shows that speech analysis in a machine learning pipeline is a promising approach for detecting suicidality among Veterans.

  Access Paper or Ask Questions

DeepSpectrumLite: A Power-Efficient Transfer Learning Framework for Embedded Speech and Audio Processing from Decentralised Data

Apr 23, 2021
Shahin Amiriparian, Tobias Hübner, Maurice Gerczuk, Sandra Ottl, Björn W. Schuller

Deep neural speech and audio processing systems have a large number of trainable parameters, a relatively complex architecture, and require a vast amount of training data and computational power. These constraints make it more challenging to integrate such systems into embedded devices and utilise them for real-time, real-world applications. We tackle these limitations by introducing DeepSpectrumLite, an open-source, lightweight transfer learning framework for on-device speech and audio recognition using pre-trained image convolutional neural networks (CNNs). The framework creates and augments Mel-spectrogram plots on-the-fly from raw audio signals which are then used to finetune specific pre-trained CNNs for the target classification task. Subsequently, the whole pipeline can be run in real-time with a mean inference lag of 242.0 ms when a DenseNet121 model is used on a consumer-grade Motorola moto e7 plus smartphone. DeepSpectrumLite operates decentralised, eliminating the need for data upload for further processing. By obtaining state-of-the-art results on a set of paralinguistics tasks, we demonstrate the suitability of the proposed transfer learning approach for embedded audio signal processing, even when data is scarce. We provide an extensive command-line interface for users and developers which is comprehensively documented and publicly available at

* 5 pages, 1 figure 

  Access Paper or Ask Questions