Presentation Attack Detection on Limited-Resource Devices Using Deep Neural Classifiers Trained on Consistent Spectrogram Fragments

The presented paper is concerned with detection of presentation attacks against unsupervised remote biometric speaker verification, using a well-known challenge–response scheme. We propose a novel approach to convolutional phoneme classifier training, which ensures high phoneme recognition accuracy even for significantly simplified network architectures, thus enabling efficient utterance verification on resource-limited hardware, such as mobile phones or embedded devices. We consider Deep Convolutional Neural Networks operating on windows of speech Mel-Spectrograms as a means for phoneme recognition, and we show that one can boost the performance of highly simplified neural architectures by modifying the principle underlying training set construction. Instead of generating training examples by slicing spectrograms using a sliding window, as it is commonly done, we propose to maximize the consistency of phoneme-related spectrogram structures that are to be learned, by choosing only spectrogram chunks from the central regions of phoneme articulation intervals. This approach enables better utilization of the limited capacity of the considered simplified networks, as it significantly reduces a within-class data scatter. We show that neural architectures comprising as few as dozens of thousands parameters can successfully—with accuracy of up to 76%, solve the 39-phoneme recognition task (we use the English language TIMIT database for experimental verification of the method). We also show that ensembling of simple classifiers, using a basic bagging method, boosts the recognition accuracy by another 2–3%, offering Phoneme Error Rates at the level of 23%, which approaches the accuracy of the state-of-the-art deep neural architectures that are one to two orders of magnitude more complex than the proposed solution. This, in turn, enables executing reliable presentation attack detection, based on just few-syllable long challenges on highly resource-limited computing hardware.

A Pronunciation Prior Assisted Vowel Reduction Detection Framework with Multi-Stream Attention Method

Vowel reduction is a common pronunciation phenomenon in stress-timed languages like English. Native speakers tend to weaken unstressed vowels into a schwa-like sound. It is an essential factor that makes the accent of language learners sound unnatural. To improve vowel reduction detection in a phoneme recognition framework, we propose an end-to-end vowel reduction detection method that introduces pronunciation prior knowledge as auxiliary information. In particular, we have designed two methods for automatically generating pronunciation prior sequences from reference texts and have implemented a main and auxiliary encoder structure that uses hierarchical attention mechanisms to utilize the pronunciation prior information and acoustic information dynamically. In addition, we also propose a method to realize the feature enhancement after encoding by using the attention mechanism between different streams to obtain expanded multi-streams. Compared with the HMM-DNN hybrid method and the general end-to-end method, the average F1 score of our approach for the two types of vowel reduction detection increased by 8.8% and 6.9%, respectively. The overall phoneme recognition rate increased by 5.8% and 5.0%, respectively. The experimental part further analyzes why the pronunciation prior knowledge auxiliary input is effective and the impact of different pronunciation prior knowledge types on performance.

Deep Learning-Based Detection of Articulatory Features in Arabic and English Speech

This study proposes using object detection techniques to recognize sequences of articulatory features (AFs) from speech utterances by treating AFs of phonemes as multi-label objects in speech spectrogram. The proposed system, called AFD-Obj, recognizes sequence of multi-label AFs in speech signal and localizes them. AFD-Obj consists of two main stages: firstly, we formulate the problem of AFs detection as an object detection problem and prepare the data to fulfill requirement of object detectors by generating a spectral three-channel image from the speech signal and creating the corresponding annotation for each utterance. Secondly, we use annotated images to train the proposed system to detect sequences of AFs and their boundaries. We test the system by feeding spectrogram images to the system, which will recognize and localize multi-label AFs. We investigated using these AFs to detect the utterance phonemes. YOLOv3-tiny detector is selected because of its real-time property and its support for multi-label detection. We test our AFD-Obj system on Arabic and English languages using KAPD and TIMIT corpora, respectively. Additionally, we propose using YOLOv3-tiny as an Arabic phoneme detection system (i.e., PD-Obj) to recognize and localize a sequence of Arabic phonemes from whole speech utterances. The proposed AFD-Obj and PD-Obj systems achieve excellent results for Arabic corpus and comparable to the state-of-the-art method for English corpus. Moreover, we showed that using only one-scale detection is suitable for AFs detection or phoneme recognition.