arXiv subjects

Ahmed Ali

Publications and source records attributed to Ahmed Ali.

At least 55 records · Page 3Linked to original sources

arXiv2021

Towards One Model to Rule All: Multilingual Strategy for Dialectal Code-Switching Arabic ASR

With the advent of globalization, there is an increasing demand for multilingual automatic speech recognition (ASR), handling language and dialectal variation of spoken content. Recent studies show its efficacy over monolingual systems. In this study, we design a large multilingual end-to-end ASR using self-attention based conformer architecture. We trained the system using Arabic (Ar), English (En) and French (Fr) languages. We evaluate the system performance handling: (i) monolingual (Ar, En and Fr); (ii) multi-dialectal (Modern Standard Arabic, along with dialectal variation such as Egyptian and Moroccan); (iii) code-switching -- cross-lingual (Ar-En/Fr) and dialectal (MSA-Egyptian dialect) test cases, and compare with current state-of-the-art systems. Furthermore, we investigate the influence of different embedding/character representations including character vs word-piece; shared vs distinct input symbol per language. Our findings demonstrate the strength of such a model by outperforming state-of-the-art monolingual dialectal Arabic and code-switching Arabic ASR.

cs.CL↗

arXiv2021

Arabic Code-Switching Speech Recognition using Monolingual Data

Code-switching in automatic speech recognition (ASR) is an important challenge due to globalization. Recent research in multilingual ASR shows potential improvement over monolingual systems. We study key issues related to multilingual modeling for ASR through a series of large-scale ASR experiments. Our innovative framework deploys a multi-graph approach in the weighted finite state transducers (WFST) framework. We compare our WFST decoding strategies with a transformer sequence to sequence system trained on the same data. Given a code-switching scenario between Arabic and English languages, our results show that the WFST decoding approaches were more suitable for the intersentential code-switching datasets. In addition, the transformer system performed better for intrasentential code-switching task. With this study, we release an artificially generated development and test sets, along with ecological code-switching test set, to benchmark the ASR performance.

cs.CL↗

arXiv2021

Arabic Speech Recognition by End-to-End, Modular Systems and Human

Recent advances in automatic speech recognition (ASR) have achieved accuracy levels comparable to human transcribers, which led researchers to debate if the machine has reached human performance. Previous work focused on the English language and modular hidden Markov model-deep neural network (HMM-DNN) systems. In this paper, we perform a comprehensive benchmarking for end-to-end transformer ASR, modular HMM-DNN ASR, and human speech recognition (HSR) on the Arabic language and its dialects. For the HSR, we evaluate linguist performance and lay-native speaker performance on a new dataset collected as a part of this study. For ASR the end-to-end work led to 12.5%, 27.5%, 33.8% WER; a new performance milestone for the MGB2, MGB3, and MGB5 challenges respectively. Our results suggest that human performance in the Arabic language is still considerably better than the machine with an absolute WER gap of 3.5% on average.

eess.AS↗

arXiv2021

QASR: QCRI Aljazeera Speech Resource -- A Large Scale Annotated Arabic Speech Corpus

We introduce the largest transcribed Arabic speech corpus, QASR, collected from the broadcast domain. This multi-dialect speech dataset contains 2,000 hours of speech sampled at 16kHz crawled from Aljazeera news channel. The dataset is released with lightly supervised transcriptions, aligned with the audio segments. Unlike previous datasets, QASR contains linguistically motivated segmentation, punctuation, speaker information among others. QASR is suitable for training and evaluating speech recognition systems, acoustics- and/or linguistics- based Arabic dialect identification, punctuation restoration, speaker identification, speaker linking, and potentially other NLP modules for spoken data. In addition to QASR transcription, we release a dataset of 130M words to aid in designing and training a better language model. We show that end-to-end automatic speech recognition trained on QASR reports a competitive word error rate compared to the previous MGB-2 corpus. We report baseline results for downstream natural language processing tasks such as named entity recognition using speech transcript. We also report the first baseline for Arabic punctuation restoration. We make the corpus available for the research community.

cs.CL↗

arXiv2020

Interpretation of LHCb Hidden-Charm Pentaquarks within the Compact Diquark Model

The LHCb collaboration have recently updated their analysis of the resonant $J/ψ\, p$ mass spectrum in the decay $Λ_b^0 \to J/ψ\, p\, K^-$, making use of their combined Run~1 and Run~2 data. In the updated analysis, three narrow states, $P_c (4312)^+$, $P_c (4440)^+$, and $P_c (4457)^+$, are observed. The spin-parity assignments of these states are not yet known. We interpret these narrow resonances as compact hidden-charm diquark-diquark-antiquark pentaquarks. Using an effective Hamiltonian, based on constituent quarks and diquarks, we calculate the pentaquark mass spectrum for the complete $SU (3)_F$ lowest $S$- and $P$-wave multiplets, taking into account dominant spin-spin, spin-orbit, orbital and tensor interactions. The resulting spectrum is very rich and we work out the quark flavor compositions, masses, and $J^P$ quantum numbers of the pentaquarks. However, heavy quark symmetry restricts the observable states in $Λ_b$-baryon, as well as in the decays of the other weakly-decaying $b$-baryons, $Ξ_b$ and $Ω_b$. In addition, some of the pentaquark states are estimated to lie below the $J/ψ\, p$ threshold in $Λ_b$-decays (and corresponding thresholds in $Ξ_b$- and $Ω_b$-decays). They decay via $c \bar c$ annihilation into light hadrons or a dilepton pair, and are expected to be narrower than the $P_c$-states observed. We anticipate their discovery, as well as of the other pentaquark states present in the spectrum at the LHC, and in the long-term future at a Tera-$Z$ factory.

hep-ph↗

arXiv2020

Tetraquark Interpretation and Production Mechanism of the Belle $Y_b (10750)$-Resonance

Recently, the Belle Collaboration has updated the analysis of the cross sections for the processes $e^+ e^- \to Υ(nS)\, π^+ π^-$ ($n = 1,\, 2,\, 3$) in the $e^+ e^-$ center-of-mass energy range from 10.52 to 11.02~GeV. A new structure, called $Y_b (10750)$, with the mass $M (Y_b) = (10752.7 \pm 5.9^{+0.7}_{-1.1})$~MeV and the Breit-Wigner width $Γ(Y_b) = (35.5^{+17.6 +3.9}_{-11.3 -3.3})$~MeV was observed. We interpret $Y_b (10750)$ as a compact $J^{PC} = 1^{--}$ state with a dominant tetraquark component. The mass eigenstate $Y_b (10750)$ is treated as a linear combination of the diquark-antidiquark and $b \bar b$ components due to the mixing via gluonic exchanges shown recently to arise in the limit of large number of quark colors. The mixing angle between $Y_b$ and $Υ(5S)$ can be estimated from the electronic width, recently determined to be $Γ_{ee} (Y_b) = (13.7 \pm 1.8)$~eV. The mixing provides a plausible mechanism for $Y_b (10750)$ production in high energy collisions from its $b \bar b$ component and we work out the Drell-Yan and prompt production cross sections for $p p \to Y_b (10750) \to Υ(nS)\, π^+ π^-$ at the LHC. The resonant part of the dipion invariant mass spectrum in $Y_b (10750) \to Υ(1S)\, π^+ π^-$ and the corresponding angular distribution of $π^+$-meson in the dipion rest frame are presented.

hep-ph↗

arXiv2020

Efficient Arabic emotion recognition using deep neural networks

Emotion recognition from speech signal based on deep learning is an active research area. Convolutional neural networks (CNNs) may be the dominant method in this area. In this paper, we implement two neural architectures to address this problem. The first architecture is an attention-based CNN-LSTM-DNN model. In this novel architecture, the convolutional layers extract salient features and the bi-directional long short-term memory (BLSTM) layers handle the sequential phenomena of the speech signal. This is followed by an attention layer, which extracts a summary vector that is fed to the fully connected dense layer (DNN), which finally connects to a softmax output layer. The second architecture is based on a deep CNN model. The results on an Arabic speech emotion recognition task show that our innovative approach can lead to significant improvements (2.2% absolute improvements) over a strong deep CNN baseline system. On the other hand, the deep CNN models are significantly faster than the attention based CNN-LSTM-DNN models in training and classification.

cs.CL↗

arXiv2020

Transverse Energy-Energy Correlations of jets in the electron-proton Deep Inelastic Scattering at HERA

We study the event shape variables, transverse energy energy correlation TEEC $(\cos ϕ)$ and its asymmetry ATEEC $(\cos ϕ)$ in deep inelastic scattering (DIS) at the electron-proton collider HERA, where $ϕ$ is the angle between two jets defined using a transverse-momentum $(k_T)$ jet algorithm. At HERA, jets are defined in the Breit frame, and the leading nontrivial transverse energy energy correlations arise from the 3-jet configurations. With the help of the NLOJET++, these functions are calculated in the leading order (LO) and the next-to-leading order (NLO) approximations in QCD at the electron-proton center-of-mass energy $\sqrt{s}=314$ GeV. We restrict the angular region to $-0.8 \leq \cos ϕ\leq 0.8$, as the forward- and backward-angular regions require resummed logarithmic corrections, which we have neglected in this work. Following experimental jet-analysis at HERA, we restrict the DIS-variables $x$, $y=Q^2/(x s)$, where $Q^2=-q^2$ is the negative of the momentum transfer squared $q^2$, to $0 \leq x \leq 1$, $0.2 \leq y \leq 0.6$, and the pseudo-rapidity variable in the laboratory frame $(η^{\rm {lab}})$ to the range $-1 \leq η^{\rm {lab}} \leq 2.5$. The TEEC and ATEEC functions are worked out for two ranges in $Q^2$, defined by $5.5~{\rm GeV}^2 \leq Q^2 \leq 80~{\rm GeV}^2$, called the low-$Q^2$-range, and $150~{\rm GeV}^2 \leq Q^2 \leq 1000~{\rm GeV}^2$, called the high-$Q^2$-range. We show the sensitivity of these functions on the parton distribution functions (PDFs), the factorization $(μ_F)$ and renormalization $(μ_R)$ scales, and on $α_s(M_Z)$. Of these the correlations are stable against varying the scale $μ_F$ and the PDFs, but they do depend on $μ_R$. These studies are useful in the analysis of the HERA data, including the determination of $α_s(M_Z)$ from the shape variables.

hep-ph↗

arXiv2020

Word Error Rate Estimation Without ASR Output: e-WER2

Measuring the performance of automatic speech recognition (ASR) systems requires manually transcribed data in order to compute the word error rate (WER), which is often time-consuming and expensive. In this paper, we continue our effort in estimating WER using acoustic, lexical and phonotactic features. Our novel approach to estimate the WER uses a multistream end-to-end architecture. We report results for systems using internal speech decoder features (glass-box), systems without speech decoder features (black-box), and for systems without having access to the ASR system (no-box). The no-box system learns joint acoustic-lexical representation from phoneme recognition results along with MFCC acoustic features to estimate WER. Considering WER per sentence, our no-box system achieves 0.56 Pearson correlation with the reference evaluation and 0.24 root mean square error (RMSE) across 1,400 sentences. The estimated overall WER by e-WER2 is 30.9% for a three hours test set, while the WER computed using the reference transcriptions was 28.5%.

eess.AS↗

arXiv2020

What Was Written vs. Who Read It: News Media Profiling Using Text Analysis and Social Media Context

Predicting the political bias and the factuality of reporting of entire news outlets are critical elements of media profiling, which is an understudied but an increasingly important research direction. The present level of proliferation of fake, biased, and propagandistic content online, has made it impossible to fact-check every single suspicious claim, either manually or automatically. Alternatively, we can profile entire news outlets and look for those that are likely to publish fake or biased content. This approach makes it possible to detect likely "fake news" the moment they are published, by simply checking the reliability of their source. From a practical perspective, political bias and factuality of reporting have a linguistic aspect but also a social context. Here, we study the impact of both, namely (i) what was written (i.e., what was published by the target medium, and how it describes itself on Twitter) vs. (ii) who read it (i.e., analyzing the readers of the target medium on Facebook, Twitter, and YouTube). We further study (iii) what was written about the target medium on Wikipedia. The evaluation results show that what was written matters most, and that putting all information sources together yields huge improvements over the current state-of-the-art.

cs.CL↗

arXiv2020

Analyzing Phonetic and Graphemic Representations in End-to-End Automatic Speech Recognition

End-to-end neural network systems for automatic speech recognition (ASR) are trained from acoustic features to text transcriptions. In contrast to modular ASR systems, which contain separately-trained components for acoustic modeling, pronunciation lexicon, and language modeling, the end-to-end paradigm is both conceptually simpler and has the potential benefit of training the entire system on the end task. However, such neural network models are more opaque: it is not clear how to interpret the role of different parts of the network and what information it learns during training. In this paper, we analyze the learned internal representations in an end-to-end ASR model. We evaluate the representation quality in terms of several classification tasks, comparing phonemes and graphemes, as well as different articulatory features. We study two languages (English and Arabic) and three datasets, finding remarkable consistency in how different properties are represented in different layers of the deep neural network.

cs.CL↗

arXiv2020

Interpretation of $Y_b (10750)$ as a tetraquark and its production mechanism

Recently, the Belle Collaboration has updated the analysis of the cross sections for the processes $e^+ e^- \to Υ(nS)\, π^+ π^-$ ($n = 1,\, 2,\, 3$) in the $e^+ e^-$ center-of-mass energy range from 10.52 to 11.02 GeV. A new structure, called here $Y_b (10750)$, with the mass $M (Y_b) = (10752.7 \pm 5.9^{+0.7}_{-1.1})$ MeV and the Breit-Wigner width $Γ(Y_b) = (35.5^{+17.6 +3.9}_{-11.3 -3.3})$ MeV was observed \cite{Abdesselam:2019gth}. We interpret $Y_b (10750)$ as a compact $J^{PC} = 1^{--}$ state with a dominant tetraquark component. The mass eigenstate $Y_b (10750)$ is treated as a linear combination of the diquark-antidiquark and $b \bar b$ components due to the mixing via gluonic exchanges shown recently to arise in the limit of large number of quark colors. The mixing angle between $Y_b$ and $Υ(5S)$ can be estimated from the electronic width, recently determined to be $Γ_{ee} (Y_b) = (13.7 \pm 1.8)$ eV. The mixing provides a plausible mechanism for $Y_b (10750)$ production in high energy collisions from its $b \bar b$ component and we work out the Drell-Yan and prompt production cross sections for $p p \to Y_b (10750) \to Υ(nS)\, π^+ π^-$ at the LHC. The resonant part of the dipion invariant mass spectrum in $Y_b (10750) \to Υ(1S)\, π^+ π^-$ and the corresponding angular distribution of $π^+$-meson in the dipion rest frame are presented as an example.

hep-ph↗

arXiv2019

Predicting the Leading Political Ideology of YouTube Channels Using Acoustic, Textual, and Metadata Information

We address the problem of predicting the leading political ideology, i.e., left-center-right bias, for YouTube channels of news media. Previous work on the problem has focused exclusively on text and on analysis of the language used, topics discussed, sentiment, and the like. In contrast, here we study videos, which yields an interesting multimodal setup. Starting with gold annotations about the leading political ideology of major world news media from Media Bias/Fact Check, we searched on YouTube to find their corresponding channels, and we downloaded a recent sample of videos from each channel. We crawled more than 1,000 YouTube hours along with the corresponding subtitles and metadata, thus producing a new multimodal dataset. We further developed a multimodal deep-learning architecture for the task. Our analysis shows that the use of acoustic signal helped to improve bias detection by more than 6% absolute over using text and metadata only. We release the dataset to the research community, hoping to help advance the field of multi-modal political bias detection.

cs.CL↗

arXiv2019

Mass spectrum of the hidden-charm pentaquarks in the compact diquark model

The LHCb collaboration have recently updated their analysis of the resonant $J/ψp$ mass spectrum in the decay $Λ_b^0 \to J/ψp K^-$, making use of their combined Run 1 and Run 2 data. In the updated analysis, three narrow states, $P_c (4312)^+$, $P_c (4440)^+$,and $P_c (4457)^+$, are observed. The spin-parity assignments of these states are not yet known. We interpret these narrow resonances as compact hidden-charm diquark-diquark-antiquark pentaquarks. Using an effective Hamiltonian, based on constituent quarks and diquarks, we calculate the pentaquark mass spectrum for the complete $SU (3)_F$ lowest $S$- and $P$-wave multiplets, taking into account dominant spin-spin, spin-orbit, orbital and tensor interactions. The resulting spectrum is very rich and we work out the quark flavor compositions, masses, and $J^P$ quantum numbers of the pentaquarks. However, heavy quark symmetry restricts the observable states in $Λ_b$-baryon, as well as in the decays of the other weakly-decaying $b$-baryons, $Ξ_b$ and $Ω_b$. In addition, some of the pentaquark states are estimated to lie below the $J/ψp$ threshold in $Λ_b$-decays (and corresponding thresholds in $Ξ_b$- and $Ω_b$-decays). They decay via $c \bar c$ annihilation into light hadrons or a dilepton pair, and are expected to be narrower than the $P_c$-states observed. We anticipate their discovery, as well as of the other pentaquark states present in the spectrum at the LHC, and in the long-term future at a Tera-$Z$ factory.

hep-ph↗

arXiv2019

Detecting Deception in Political Debates Using Acoustic and Textual Features

We present work on deception detection, where, given a spoken claim, we aim to predict its factuality. While previous work in the speech community has relied on recordings from staged setups where people were asked to tell the truth or to lie and their statements were recorded, here we use real-world political debates. Thanks to the efforts of fact-checking organizations, it is possible to obtain annotations for statements in the context of a political discourse as true, half-true, or false. Starting with such data from the CLEF-2018 CheckThat! Lab, which was limited to text, we performed alignment to the corresponding videos, thus producing a multimodal dataset. We further developed a multimodal deep-learning architecture for the task of deception detection, which yielded sizable improvements over the state of the art for the CLEF-2018 Lab task 2. Our experiments show that the use of the acoustic signal consistently helped to improve the performance compared to using textual and metadata features only, based on several different evaluation measures. We release the new dataset to the research community, hoping to help advance the overall field of multimodal deception detection.

cs.CL↗

arXiv2019

DARTS: Dialectal Arabic Transcription System

We present the speech to text transcription system, called DARTS, for low resource Egyptian Arabic dialect. We analyze the following; transfer learning from high resource broadcast domain to low-resource dialectal domain and semi-supervised learning where we use in-domain unlabeled audio data collected from YouTube. Key features of our system are: A deep neural network acoustic model that consists of a front end Convolutional Neural Network (CNN) followed by several layers of Time Delayed Neural Network (TDNN) and Long-Short Term Memory Recurrent Neural Network (LSTM); sequence discriminative training of the acoustic model; n-gram and recurrent neural network language model for decoding and N-best list rescoring. We show that a simple transfer learning method can achieve good results. The results are further improved by using unlabeled data from YouTube in a semi-supervised setup. Various systems are combined to give the final system that achieves the lowest word error on on the community standard Egyptian-Arabic speech dataset (MGB-3).

cs.CL↗

arXiv2019

The MGB-2 Challenge: Arabic Multi-Dialect Broadcast Media Recognition

This paper describes the Arabic Multi-Genre Broadcast (MGB-2) Challenge for SLT-2016. Unlike last year's English MGB Challenge, which focused on recognition of diverse TV genres, this year, the challenge has an emphasis on handling the diversity in dialect in Arabic speech. Audio data comes from 19 distinct programmes from the Aljazeera Arabic TV channel between March 2005 and December 2015. Programmes are split into three groups: conversations, interviews, and reports. A total of 1,200 hours have been released with lightly supervised transcriptions for the acoustic modelling. For language modelling, we made available over 110M words crawled from Aljazeera Arabic website Aljazeera.net for a 10 year duration 2000-2011. Two lexicons have been provided, one phoneme based and one grapheme based. Finally, two tasks were proposed for this year's challenge: standard speech transcription, and word alignment. This paper describes the task data and evaluation process used in the MGB challenge, and summarises the results obtained.

cs.CL↗

arXiv2019

Domain Attentive Fusion for End-to-end Dialect Identification with Unknown Target Domain

End-to-end deep learning language or dialect identification systems operate on the spectrogram or other acoustic feature and directly generate identification scores for each class. An important issue for end-to-end systems is to have some knowledge of the application domain, because the system can be vulnerable to use cases that were not seen in the training phase; such a scenario is often referred to as a domain mismatched condition. In general, we assume that there is enough variation in the training dataset to expose the system to multiple domains. In this work, we study how to best make use a training dataset in order to have maximum effectiveness on unknown target domains. Our goal is to process the input without any knowledge of the target domain while preserving robust performance on other domains as well. To accomplish this objective, we propose a domain attentive fusion approach for end-to-end dialect/language identification systems. To help with experimentation, we collect a dataset from three different domains, and create experimental protocols for a domain mismatched condition. The results of our proposed approach, which were tested on a variety of broadcast and YouTube data, shows significant performance gain compared to traditional approaches, even without any prior target domain information.

eess.AS↗