Speech Recognition Arabic

This paper presents the application of morpheme-based and factored language models in an Amharic speech recognition task. Since using morphemes in both acoustic and language models results, mostly, in performance degradation due to acoustic confusability and since it is problematic to use factored language models in standard word decoders, we applied the models in a lattice rescoring framework. Lattices of 100 best alternatives for each test sentence of the 5k development test set have been generated using a baseline speech recognizer that uses a word-based backoff bigram language model. The lattices have then been rescored with various morpheme-based and factored language models and a slight improvement in word recognition accuracy has been observed.

The Arabic language has many different dialects and it must be recognized before using the automatic speech recognition (ASR). On the other hand, it is observed in all Arab countries that the standard Arabic language is widely written and used in an official speech, newspapers, public administration, and schools but it is not used in daily conversations instead the dialect is widely spoken in daily life and rarely written. In this paper, we examine the difficult task of properly identifying various Arabic dialects and propose a system developed to identify a set of four regional and modern standard Arabic speeches, based on speech recognition using Hidden Markov Models (HMMs) algorithms. HMMs have become a very popular way to build a speech recognition system. It is set as hidden states and possibilities of transition from one state to another. Due to the similarities and differences between the Arabic dialects, speeches collected from the ADI5 datasets were retrieved from the MGB-3...

Development of Multilingual Automatic Speech Recognition (ASR) systems enables to share existing speech and text corpora among languages. We have conducted experiments on the development of multilingual Acoustic Models (AM) and Language Models (LM) for Tigrigna. Using Amharic Deep Neural Network (DNN) AM, Tigrigna pronunciation dictionary and trigram LM, we achieved a Word Error Rate (WER) of 30.9% for Tigrigna. Adding training speech from the target language (Tigrigna) to the whole training speech of the donor language (Amharic) continuously reduces WER with the amount of added data. We have also developed different (including recurrent neural networks based) multilingual LMs and achieved a relative WER reduction of 3.56% compared to the use of monolingual trigram LMs. Considering scarcity of computational resources to decode with very large vocabularies, we have also experimented on the use of morphemes as pronunciation and language modeling units. We have achieved character error rate (CER) of 7.9% which is relatively lower by 38.3% to 1.3% than the CER of the word-based models of smaller vocabularies than 162k. Our results show the possibility of developing ASR system for an Ethio-Semitic language using an existing speech and text corpora of another language in the family.

This paper presents the application of morpheme-based and factored language models in an Amharic speech recognition task. Since using morphemes in both acoustic and language models results, mostly, in performance degradation due to acoustic confusability and since it is problematic to use factored language models in standard word decoders, we applied the models in a lattice rescoring framework. Lattices of 100 best alternatives for each test sentence of the 5k development test set have been generated using a baseline speech recognizer that uses a word-based backoff bigram language model. The lattices have then been rescored with various morpheme-based and factored language models and a slight improvement in word recognition accuracy has been observed.

In this paper, we introduce the first large vocabulary speech recognition system (LVSR) for the Central Kurdish language, named Jira. The Kurdish language is an Indo-European language spoken by more than 30 million people in several countries, but due to the lack of speech and text resources, there is no speech recognition system for this language. To fill this gap, we introduce the first speech corpus and pronunciation lexicon for the Kurdish language. Regarding speech corpus, we designed a sentence collection in which the ratio of di-phones in the collection resembles the real data of the Central Kurdish language. The designed sentences are uttered by 576 speakers in a controlled environment with noise-free microphones (called AsoSoft Speech-Office) and in Telegram social network environment using mobile phones (denoted as AsoSoft Speech-Crowdsourcing), resulted in 43.68 hours of speech. Besides, a test set including 11 different document topics is designed and recorded in two corresponding speech conditions (i.e., Office and Crowdsourcing). Furthermore, a 60K pronunciation lexicon is prepared in this research in which we faced several challenges and proposed solutions for them. The Kurdish language has several dialects and sub-dialects that results in many lexical variations. Our methods for script standardization of lexical variations and automatic pronunciation of the lexicon tokens are presented in detail. To setup the recognition engine, we used the Kaldi toolkit. A statistical tri-gram language model that is extracted from the AsoSoft text corpus is used in the system. Several standard recipes including HMM-based models (i.e., mono, tri1, tr2, tri2, tri3), SGMM, and DNN methods are used to generate the acoustic model. These methods are trained with AsoSoft Speech-Office and AsoSoft Speech-Crowdsourcing and a combination of them. The best performance achieved by the SGMM acoustic model which results in 13.9% of the average word error rate (on different document topics) and 4.9% for the general topic.

The Donate Speech campaign has so far succeeded in gathering approximately 3600 h of ordinary, colloquial Finnish speech into the Lahjoita puhetta (Donate Speech) corpus. The corpus includes over twenty thousand speakers from all the regions of Finland and from all age brackets. The primary goals of the collection were to create a representative, large-scale resource to study spontaneous spoken Finnish and to accelerate the development of language technology and speech-based services. In this paper, we present the collection process and the collected corpus, and showcase its versatility through multiple use cases. The evaluated use cases include: automatic speech recognition of spontaneous speech, detection of age, gender, dialect and topic and metadata analysis. We provide benchmarks for the use cases, as well downloadable, trained baseline systems with open-source code for reproducibility. One further use case is to verify the metadata and transcripts given in this corpus itself, ...

Speech recognition systems play an important role in human-machine interactions. Many systems exist for Arabic speech, however, there are limited systems for dialectal Arabic speech. The Arabic language comprises many properties, some of which are ideal for building automatic speech recognition systems such as syntax and phonology, while other properties are unsuitable for developing speech systems. Importantly, most data are in non-diacritized form, vary in dialect, and contain morphological complexity. Moreover, the Arabic dialects lack a standard structure. In this paper, we highlighted the works and frameworks that have been developed in the last fourteen years of dialectal Arabic speech recognition systems. The paper also presented an analysis and evaluation for several studies using different approaches and techniques. The main goal of this paper is to compare and discuss different studies in dialectal Arabic speech systems including several criteria such as techniques, datasets, evaluation metrics, and dialect types. The study also includes a description of some techniques used in all steps of the dialectal Arabic speech system such as hidden Markov models (HMM), convolutional neural network (CNN), and deep neural network (DNN). In addition, we introduced the challenges and problems of dialectal Arabic speech recognition systems. Overall, more studies are required to obtain a more accurate speech system for dialectal Arabic.

Speech recognition basically means talking to a computer, having it recognize what Speakers are saying. Speech is common and efficient form of communication method for people to interact with each other. The person would like to interact with computer via speech. It can be accomplished by speech recognition system in which computer identifies the word spoken by a speaker into a microphone. Speech recognition is becoming more complex and a challenging task. The research is focusing on large vocabulary, continuous speech capabilities and speaker independence. This paper reviews methods and technologies available for ASR process

Towards developing high-performing ASR for low-resource languages, approaches to address the lack of resources are to make use of data from multiple languages, and to augment the training data by creating acoustic variations. In this work we present a single grapheme-based ASR model learned on 7 geographically proximal languages, using standard hybrid BLSTM-HMM acoustic models with lattice-free MMI objective. We build the single ASR grapheme set via taking the union over each language-specific grapheme set, and we find such multilingual graphemic hybrid ASR model can perform language-independent recognition on all 7 languages, and substantially outperform each monolingual ASR model. Secondly, we evaluate the efficacy of multiple data augmentation alternatives within language, as well as their complementarity with multilingual modeling. Overall, we show that the proposed multilingual graphemic hybrid ASR with various data augmentation can not only recognize any within training set la...

Speech recognition systems play an important role in human-machine interactions. Many systems exist for Arabic speech, however, there are limited systems for dialectal Arabic speech. The Arabic language comprises many properties, some of which are ideal for building automatic speech recognition systems such as syntax and phonology, while other properties are unsuitable for developing speech systems. Importantly, most data are in non-diacritized form, vary in dialect, and contain morphological complexity. Moreover, the Arabic dialects lack a standard structure. In this paper, we highlighted the works and frameworks that have been developed in the last fourteen years of dialectal Arabic speech recognition systems. The paper also presented an analysis and evaluation for several studies using different approaches and techniques. The main goal of this paper is to compare and discuss different studies in dialectal Arabic speech systems including several criteria such as techniques, datasets, evaluation metrics, and dialect types. The study also includes a description of some techniques used in all steps of the dialectal Arabic speech system such as hidden Markov models (HMM), convolutional neural network (CNN), and deep neural network (DNN). In addition, we introduced the challenges and problems of dialectal Arabic speech recognition systems. Overall, more studies are required to obtain a more accurate speech system for dialectal Arabic.

End-to-end automatic speech recognition (ASR) has simplified the traditional ASR system building pipeline by eliminating the need to have multiple components and also the requirement for expert linguistic knowledge for creating pronunciation dictionaries. Therefore, end-to-end ASR fits well when building systems for new domains. However, one major drawback of end-to-end ASR is that, it is necessary to have a larger amount of labeled speech in comparison to traditional methods. Therefore, in this paper, we explore domain adaptation approaches for end-to-end ASR in low-resource settings. We show that joint domain identification and speech recognition by inserting a symbol for domain at the beginning of the label sequence, factorized hidden layer adaptation and a domain-specific gating mechanism improve the performance for a low-resource target domain. Furthermore, we also show the robustness of proposed adaptation methods to an unseen domain, when only 3 hours of untranscribed data is available with improvements reporting upto 8.7% relative.

Speech recognition systems play an important role in human-machine interactions. Many systems exist for Arabic speech, however, there are limited systems for dialectal Arabic speech. The Arabic language comprises many properties, some of which are ideal for building automatic speech recognition systems such as syntax and phonology, while other properties are unsuitable for developing speech systems. Importantly, most data are in non-diacritized form, vary in dialect, and contain morphological complexity. Moreover, the Arabic dialects lack a standard structure. In this paper, we highlighted the works and frameworks that have been developed in the last fourteen years of dialectal Arabic speech recognition systems. The paper also presented an analysis and evaluation for several studies using different approaches and techniques. The main goal of this paper is to compare and discuss different studies in dialectal Arabic speech systems including several criteria such as techniques, datasets, evaluation metrics, and dialect types. The study also includes a description of some techniques used in all steps of the dialectal Arabic speech system such as hidden Markov models (HMM), convolutional neural network (CNN), and deep neural network (DNN). In addition, we introduced the challenges and problems of dialectal Arabic speech recognition systems. Overall, more studies are required to obtain a more accurate speech system for dialectal Arabic.

Recognition of Hungarian conversational telephone speech is challenging due to the informal style and morphological richness of the language. Recurrent Neural Network Language Model (RNNLM) can provide remedy for the high perplexity of the task; however, twopass decoding introduces a considerable processing delay. In order to eliminate this delay we investigate approaches aiming at the complexity reduction of RNNLM, while preserving its accuracy. We compare the performance of conventional back-off n-gram language models (BNLM), BNLM approximation of RNNLMs (RNN-BNLM) and RNN n-grams in terms of perplexity and word error rate (WER). Morphological richness is often addressed by using statistically derived subwords-morphs-in the language models, hence our investigations are extended to morphbased models, as well. We found that using RNN-BNLMs 40% of the RNNLM perplexity reduction can be recovered, which is roughly equal to the performance of a RNN 4-gram model. Combining morph-based modeling and approximation of RNNLM, we were able to achieve 8% relative WER reduction and preserve real-time operation of our conversational telephone speech recognition system.

Hybrid deep neural network hidden Markov models (DNN-HMM) have achieved impressive results on large vocabulary continuous speech recognition (LVCSR) tasks. However, the recent approaches using DNN-HMM models are not explored much for text recognition. Inspired by the current work in automatic speech recognition (ASR) and machine translation, we present an open vocabulary sub-word text recognition system. The sub-word lexicon and sub-word language model (LM) helps in overcoming the challenge of recognizing out of vocabulary (OOV) words, and a time delay neural network (TDNN) and convolution neural network (CNN) based DNN-HMM optical model (OM) efficiently models the sequence dependency in the line image. We present results on 12 datasets with training data varying from 6k lines to 600k lines. The system is built for 8 languages, i.e., English, French, Arabic, Chinese, Farsi, Tamil, Russian, and Korean. We report competitive results on several commonly used handwritten and printed text datasets.

One of the current research areas is speech recognition by aiding in the recognition of speech signals through computer applications. In this research paper, Acoustic Nudging, (AN) Model is used in re-formulating the persistence automatic speech recognition (ASR) errors that involves user’s acoustic irrational behavior which alters speech recognition accuracy. GMM helped in addressing low-resourced attribute of Yorùbá language to achieve better accuracy and system performance. From the simulated results given, it is observed that proposed Acoustic Nudging-based Gaussian Mixture Model (ANGM) improves accuracy and system performance which is evaluated based on Word Recognition Rate (WRR) and Word Error Rate (WER)given by validation accuracy, testing accuracy, and training accuracy. The evaluation results for the mean WRR accuracy achieved for the ANGM model is 95.277% and the mean Word Error Rate (WER) is 4.723%when compared to existing models. This approach thereby reduce error rate ...

We propose a method to extend a phoneme set by using a large amount of broadcast data to improve the performance of Korean spontaneous speech recognition. In the proposed method, we first extract variable-length phoneme-level segments from broadcast data and then convert them into fixed-length embedding vectors based on a long short-term memory architecture. We use decision tree-based clustering to find acoustically similar embedding vectors and then build new acoustic subword units by gathering the clustered vectors. To update the lexicon of a speech recognizer, we build a lookup table between the tri-phone units and the units derived from the decision tree. Finally, the proposed lexicon is obtained by updating the original phoneme-based lexicon by referencing the lookup table. To verify the performance of the proposed unit, we compare the proposed unit with the previous units obtained by using the segment-based k-means clustering method or the frame-based decision-tree clustering method. As a result, the proposed unit is shown to produce better performance than the previous units in both spontaneous, and read Korean speech recognition tasks.

One of the current research areas is speech recognition by aiding in the recognition of speech signals through computer applications. In this research paper, Acoustic Nudging, (AN) Model is used in re-formulating the persistence automatic speech recognition (ASR) errors that involves user’s acoustic irrational behavior which alters speech recognition accuracy. GMM helped in addressing low-resourced attribute of Yorùbá language to achieve better accuracy and system performance. From the simulated results given, it is observed that proposed Acoustic Nudging-based Gaussian Mixture Model (ANGM) improves accuracy and system performance which is evaluated based on Word Recognition Rate (WRR) and Word Error Rate (WER)given by validation accuracy, testing accuracy, and training accuracy. The evaluation results for the mean WRR accuracy achieved for the ANGM model is 95.277% and the mean Word Error Rate (WER) is 4.723%when compared to existing models. This approach thereby reduce error rate ...

in some languages like Classical Arabic (The language of the Holy Quran), phoneme duration is considered as a distinguishing cue in Quranic phonology. Phonological variation of phonemes occurrences contributes to an inaccurate pronunciation of phonemes and therefore inaccurate ASR system. Thus a good phonemes duration modeling can be an essential issue. Currently, the most effective models used in automatic speech recognition (ASR) systems are based on statistical approaches namely Hidden Markov Model (HMM). In standard HMM speech recognition framework, the duration information is poorly employed. However, previous studies have demonstrated that using an HMM with explicit duration modeling techniques have improved the recognition performances in many targeted languages. This paper presents an important phase of our ongoing work which aims to build an accurate Arabic recognizer for teaching and learning purposes. It presents an implementation of an HsMM model (Hidden semi-Markov Model) whose main role is enhancing the classical HMM duration behavior. In this model, both Gamma and Gaussian distributions are used for modeling state durations and compared with the standard geometrical distribution. Experiments have been conducted on a particular database of ten speakers and more than eight hours of speech collected from recitations of the Holy Quran in which all classical Arabic sounds are covered. Results show an accuracy improvement of about 1% over the baseline HMM-based recognizer, which proves the suitability of Gamma distribution in state duration modeling.

This paper presents the application of morpheme-based and factored language models in an Amharic speech recognition task. Since using morphemes in both acoustic and language models results, mostly, in performance degradation due to acoustic confusability and since it is problematic to use factored language models in standard word decoders, we applied the models in a lattice rescoring framework. Lattices of 100 best alternatives for each test sentence of the 5k development test set have been generated using a baseline speech recognizer that uses a word-based backoff bigram language model. The lattices have then been rescored with various morpheme-based and factored language models and a slight improvement in word recognition accuracy has been observed.

Development of Multilingual Automatic Speech Recognition (ASR) systems enables to share existing speech and text corpora among languages. We have conducted experiments on the development of multilingual Acoustic Models (AM) and Language Models (LM) for Tigrigna. Using Amharic Deep Neural Network (DNN) AM, Tigrigna pronunciation dictionary and trigram LM, we achieved a Word Error Rate (WER) of 30.9% for Tigrigna. Adding training speech from the target language (Tigrigna) to the whole training speech of the donor language (Amharic) continuously reduces WER with the amount of added data. We have also developed different (including recurrent neural networks based) multilingual LMs and achieved a relative WER reduction of 3.56% compared to the use of monolingual trigram LMs. Considering scarcity of computational resources to decode with very large vocabularies, we have also experimented on the use of morphemes as pronunciation and language modeling units. We have achieved character error rate (CER) of 7.9% which is relatively lower by 38.3% to 1.3% than the CER of the word-based models of smaller vocabularies than 162k. Our results show the possibility of developing ASR system for an Ethio-Semitic language using an existing speech and text corpora of another language in the family.

This paper presents the application of morpheme-based and factored language models in an Amharic speech recognition task. Since using morphemes in both acoustic and language models results, mostly, in performance degradation due to acoustic confusability and since it is problematic to use factored language models in standard word decoders, we applied the models in a lattice rescoring framework. Lattices of 100 best alternatives for each test sentence of the 5k development test set have been generated using a baseline speech recognizer that uses a word-based backoff bigram language model. The lattices have then been rescored with various morpheme-based and factored language models and a slight improvement in word recognition accuracy has been observed.

This paper presents the application of morpheme-based and factored language models in an Amharic speech recognition task. Since using morphemes in both acoustic and language models results, mostly, in performance degradation due to acoustic confusability and since it is problematic to use factored language models in standard word decoders, we applied the models in a lattice rescoring framework. Lattices of 100 best alternatives for each test sentence of the 5k development test set have been generated using a baseline speech recognizer that uses a word-based backoff bigram language model. The lattices have then been rescored with various morpheme-based and factored language models and a slight improvement in word recognition accuracy has been observed.

One of the current research areas is speech recognition by aiding in the recognition of speech signals through computer applications. In this research paper, Acoustic Nudging, (AN) Model is used in re-formulating the persistence automatic speech recognition (ASR) errors that involves user’s acoustic irrational behavior which alters speech recognition accuracy. GMM helped in addressing low-resourced attribute of Yorùbá language to achieve better accuracy and system performance. From the simulated results given, it is observed that proposed Acoustic Nudging-based Gaussian Mixture Model (ANGM) improves accuracy and system performance which is evaluated based on Word Recognition Rate (WRR) and Word Error Rate (WER)given by validation accuracy, testing accuracy, and training accuracy. The evaluation results for the mean WRR accuracy achieved for the ANGM model is 95.277% and the mean Word Error Rate (WER) is 4.723%when compared to existing models. This approach thereby reduce error rate ...

Voice control is a prominent interaction method on personal computing devices. While automatic speech recognition (ASR) systems are readily applicable for large audiences, there is room for further adaptation at the edge, ie. locally on devices, targeted for individual users. In this work, we explore improving ASR systems over time through a user's own interactions. Our online learning approach for speaker-adaptive language modeling leverages a user's most recent utterances to enhance the speaker dependent features and traits. We experiment with the Large-Vocabulary Continuous Speech Recognition corpus Tedlium v2, and demonstrate an average reduction in perplexity (PPL) of 19.18% and average relative reduction in word error rate (WER) of 2.80% compared to a state-of-the-art baseline on Tedlium v2.

Speech recognition has become increasingly popular in radiology reporting in the last decade. However, developing a speech recognition system for a new language in a highly specific domain requires a lot of resources, expert knowledge and skills. Therefore, commercial vendors do not offer ready-made radiology speech recognition systems for less-resourced languages. This paper describes the implementation of a radiology speech recognition system for Estonian, a language with less than one million native speakers. The system was developed in partnership with a hospital that provided a corpus of written reports for language modeling purposes. Rewrite rules for pre-processing training texts and postprocessing recognition results were created manually based on a small parallel corpus created by the hospital's radiologists, using the Thrax toolkit. Deep neural network based acoustic models were trained based on 216 hours of out-of-domain data and adapted on 14 hours of spoken radiology data, using the Kaldi toolkit. The current word error rate of the system is 5.4%. The system is in active use in real clinical environment.

This paper gives an introduction of speech recognition systems for human-computer interaction using Vietnamese language. First, the paper investigates the two most common speech recognition toolkits currently used, HTK and Sphinx, and apply these tools into Vietnamese speech recognition. Then, only HTK is selected to design an application that control a web browser (Google Chrome) and an application that control a robot. The results obtained give a comparison between the uses of the two tool kits for Vietnamese speech recognition. 

In this paper it is presented a solution for replacing the current endoscopic exams control mechanisms. This kind of exams require the gastroenterologist to perform a complex procedure, using both hands simultaneously, to manipulate the endoscope's buttons and using the foot to press a pedal in order to perform simple tasks such as capturing frames. The last procedure cannot be accomplished in real-time because the gastroenterologist needs to press an additional programmable button on the endoscope to freeze the image and then press the pedal to capture and save the frame. The presented solution replaces the pedal with a hands-free voice control module and it is capable of running on the background continuously without human physical intervention. This system was designed to be used seamlessly with the MyEndoscopy system that is being tested in some healthcare institution and uses the PocketSphinx libraries to perform real-time recognition of a small vocabulary in two different languages, namely English and Portuguese.

Phonetic dictionaries are essential components of large-vocabulary natural language speakerindependent speech recognition systems. This paper presents a rule-based technique to generate Arabic phonetic dictionaries for a large vocabulary speech recognition system. The system used classic Arabic pronunciation rules, common pronunciation rules of Modern Standard Arabic, as well as morphologically driven rules. The paper gives in detail an explanation of these rules as well as their formal mathematical presentation. The rules were used to generate a dictionary for a 5.4 hours corpus of broadcast news. The phonetic dictionary contains 23,841 definitions corresponding to about 14232 words. The generated dictionary was evaluated on an actual Arabic speech recognition system. The pronunciation rules and the phone set were validated by test cases. The Arabic speech recognition system achieves word error rate of %11.71 for fully diacritized transcription of about 1.1 hours of Arabic broadcast news.

This paper gives an introduction of speech recognition systems for human-computer interaction using Vietnamese language. First, the paper investigates the two most common speech recognition toolkits currently used, HTK and Sphinx, and apply these tools into Vietnamese speech recognition. Then, only HTK is selected to design an application that control a web browser (Google Chrome) and an application that control a robot. The results obtained give a comparison between the uses of the two tool kits for Vietnamese speech recognition. 

This paper gives an introduction of speech recognition systems for human-computer interaction using Vietnamese language. First, the paper investigates the two most common speech recognition toolkits currently used, HTK and Sphinx, and apply these tools into Vietnamese speech recognition. Then, only HTK is selected to design an application that control a web browser (Google Chrome) and an application that control a robot. The results obtained give a comparison between the uses of the two tool kits for Vietnamese speech recognition. 

In this paper it is presented a solution for replacing the current endoscopic exams control mechanisms. This kind of exams require the gastroenterologist to perform a complex procedure, using both hands simultaneously, to manipulate the endoscope's buttons and using the foot to press a pedal in order to perform simple tasks such as capturing frames. The last procedure cannot be accomplished in real-time because the gastroenterologist needs to press an additional programmable button on the endoscope to freeze the image and then press the pedal to capture and save the frame. The presented solution replaces the pedal with a hands-free voice control module and it is capable of running on the background continuously without human physical intervention. This system was designed to be used seamlessly with the MyEndoscopy system that is being tested in some healthcare institution and uses the PocketSphinx libraries to perform real-time recognition of a small vocabulary in two different languages, namely English and Portuguese.

This paper describes a system for performing alignment of subtitles to audio on multigenre broadcasts using a lightly supervised approach. Accurate alignment of subtitles plays a substantial role in the daily work of media companies and currently still requires large human effort. Here, a comprehensive approach to performing this task in an automated way using lightly supervised alignment is proposed. The paper explores the different alternatives to speech segmentation, lightly supervised speech recognition and alignment of text streams. The proposed system uses lightly supervised decoding to improve the alignment accuracy by performing language model adaptation using the target subtitles. The system thus built achieves the third best reported result in the alignment of broadcast subtitles in the Multi-Genre Broadcast (MGB) challenge, with an F1 score of 88.8%. This system is available for research and other non-commercial purposes through webASR, the University of Sheffield's cloud-based speech technology web service. Taking as inputs an audio file and untimed subtitles, webASR can produce timed subtitles in multiple formats, including TTML, WebVTT and SRT.

In this paper, various methodologies of acoustic and language models, as well as labeling methods for automatic speech recognition for spoken dialogues in emergency call centers were investigated and comparatively analyzed. Because of the fact that dialogue speech in call centers has specific context and noisy, emotional environments, available speech recognition systems show poor performance. Therefore, in order to accurately recognize dialogue speeches, the main modules of speech recognition systems—language models and acoustic training methodologies—as well as symmetric data labeling approaches have been investigated and analyzed. To find an effective acoustic model for dialogue data, different types of Gaussian Mixture Model/Hidden Markov Model (GMM/HMM) and Deep Neural Network/Hidden Markov Model (DNN/HMM) methodologies were trained and compared. Additionally, effective language models for dialogue systems were defined based on extrinsic and intrinsic methods. Lastly, our sugge...

Sphinx-4 is a flexible, modular and pluggable framework to help foster new innovations in the core research of hidden Markov model (HMM) speech recognition systems. The design of Sphinx-4 is based on patterns that have emerged from the design of past systems as well ...

Phonetic dictionaries are essential components of large-vocabulary natural language speakerindependent speech recognition systems. This paper presents a rule-based technique to generate Arabic phonetic dictionaries for a large vocabulary speech recognition system. The system used classic Arabic pronunciation rules, common pronunciation rules of Modern Standard Arabic, as well as morphologically driven rules. The paper gives in detail an explanation of these rules as well as their formal mathematical presentation. The rules were used to generate a dictionary for a 5.4 hours corpus of broadcast news. The phonetic dictionary contains 23,841 definitions corresponding to about 14232 words. The generated dictionary was evaluated on an actual Arabic speech recognition system. The pronunciation rules and the phone set were validated by test cases. The Arabic speech recognition system achieves word error rate of %11.71 for fully diacritized transcription of about 1.1 hours of Arabic broadcast news.

Deep Learning approaches have been widely known to perform better than statistical approaches. This is the first effort to investigate Recurrent Neural Network-based modeling for Punjabi speech corpus. We propose the Lattice Rescoring based RNNLM approach using the Kaldi toolkit. Experiments on single sentences showed that the Neural network-based approach performs better than n-gram based modeling approaches. A performance improvement of 7-9% on word error rate (WER) was observed on top of the state-of-the-art Punjabi speech recognition system. Index Terms-automatic speech recognition, recurrent neural network language modeling, lattice rescoring, Punjabi ASR

The Arabic language has many features such as the phonology and the syntax that make it an easy language for developing automatic speech recognition systems. Many standard techniques for acoustic and language modeling such as context dependent acoustic models and n-gram language models can be easily applied to Arabic. Some aspects of the Arabic language such as the nearly one-to-one letter-to-phone correspondence make the construction of the pronunciation lexicon even easier than in other languages. The most difficult challenges in developing speech recognition systems for Arabic are the dominance of non-diacritized text material, the several dialects, and the morphological complexity. In this chapter, we review the efforts that have been done to handle the challenges of the Arabic language for developing automatic speech recognition systems. This includes methods for automatic generation for the diacritics of the Arabic text and word pronunciation disambiguation. We also review the used approaches for handling the limited speech and text resources of the different Arabic dialects. Finally, we review the approaches used to deal with the high degree of affixation, derivation that contributes to the explosion of different word forms in Arabic.

— Speech recognition is widely researched topic around the world. It is a process of conversion of speech to text. Many scientists and researchers are busy with doing works to increase the performance of speech recognition systems. Most of the languages in the world have speech recognizer of its own. But in our mother tongue Bangla there is no working speech recognizer. This work is little try to build a Bengali speech recognizer to enrich our language. In this paper we have proposed a noble approach to develop an automatic Bangla Real Number recognizer and analyze the performance of this recognition system using the most popular speech recognizer API CMU Sphinx 4 and a popular Bangla Unicode based writing software called Avro.

This paper addresses the problem of open vocabulary automatic speech recognition (ASR) using hybrid statistical language models (LMs). Hybrid LMs differ from closed vocabulary LMs in that the word level lexicon is augmented with an inventory of sub-lexical units (SLUs). The procedures used for selecting these SLUs and expanding out-of-vocabulary (OOV) words according to the SLUs is presented in the paper. The open-vocabulary ASR performance obtained using these techniques is presented for a lecture speech task domain.

—Speech Recognition plays a very important role in day to day life. Speech Recognition is widely used and addicted by this world as it allows users to communicate with computers by recognizing their spoken language. Communication with Speech Recognition made our lives easy. There are many types of open source Speech Recognition Engine. Different types of application are built using Speech Recognition Engine. An Application is built for Air Traffic Controller. A new Software library with good grammar is proposed according to Air Traffic Controller commands. This Software library is used to build successful working application.

Sphinx-4 is a flexible, modular and pluggable framework to help foster new innovations in the core research of hidden Markov model (HMM) speech recognition systems. The design of Sphinx-4 is based on patterns that have emerged from the design of past systems as well ...

In this paper we present the creation of an Arabic version of Automated Speech Recognition System (ASR). This system is based on the open source Sphinx-4, from the Carnegie Mellon University. Which is a speech recognition system based on discrete hidden Markov models (HMMs). We investigate the changes that must be made to the model to adapt Arabic voice recognition.

The goal of this article is to provide and present information about the training procedure SpinxTrain and its eligible modifications to get accurate and robust speech recognition in a mobile GSM environment. Some modifications are based on effective preprocessing of input data in combination with the optimal setting of the number of states per model, through the adjustment of the number of tied states or number of Gaussian mixtures. Another source of increased recognition rate is the 'optimal' setting of the speech decoder. As it is a non-linear, mathematically not well tractable task containing both real and integer values, methods of evolution strategies can be successfully used (an 18.6% improvement in WER was observed compared to the original setting). All experiments and results were obtained for the Slovak speech database Mobildat, which contains recordings of 1100 speakers. The Sphinx4 recognition system was used for evaluation of the trained model. Biographical notes: Juraj Vojtko, born in 1981, received MSc in telecommunications from the Slovak University of Technology in Bratislava, Slovakia in 2005. Since 2009 he has occupied an assistant professor position at the Institute of Telecommunications at the Faculty of Electrical Engineering and Information Technology of the Slovak University of Technology in Bratislava. He has also worked as developer in commercial segment in the area of communication and information systems since 2002. The field of his research focused on speech processing specifically speech recognition and speaker identification and verification. Juraj Kačur, born in Bratislava in 1976. Master of Science degree obtained in the year 2000 and PhD in 2005 at the Faculty of Electrical Engineering and Information Technology of the Slovak university of technology (FEI STU) Bratislava. Since 2001, he occupies an assistant professor position at the institute of telecommunication at FEI STU Bratislava. Between years 2000 and 2001, he was with the Slovak Academy of Science, department of speech analysis and synthesis where he participated on several projects. The field of his research activities includes: signal processing, speech processing, speech recognition, speech detection, speaker identification, High order statistic, Wavelet transform, machine learning, ANN and HMM. Gregor Rozinaj (M'97) received MSc and PhD in telecommunications from Slovak University of Technology, Bratislava, Slovakia in 1981 and 1990, respectively. He has been a lecturer at the Institute of Telecommunications of the Slovak University of Technology since 1981. From 1992-1994, he worked on the research project devoted to speech recognition at Alcatel Research Center in Stuttgart, Germany. From 1994-1996, he was employed as a researcher at the University of Stuttgart, Germany working on a research project for automatic ship control. Since 1997, he has been a Head of the DSP group at the Institute of Telecommunications of the Slovak University of Technology, Bratislava. Since 1998, he has been an Associate Professor at the same institute. He is an author of 3 US and European patents on digital speech recognition and 1 Czechoslovak patent on fast algorithms for DSP.

Phonetic dictionaries are essential components of large-vocabulary natural language speakerindependent speech recognition systems. This paper presents a rule-based technique to generate Arabic phonetic dictionaries for a large vocabulary speech recognition system. The system used classic Arabic pronunciation rules, common pronunciation rules of Modern Standard Arabic, as well as morphologically driven rules. The paper gives in detail an explanation of these rules as well as their formal mathematical presentation. The rules were used to generate a dictionary for a 5.4 hours corpus of broadcast news. The phonetic dictionary contains 23,841 definitions corresponding to about 14232 words. The generated dictionary was evaluated on an actual Arabic speech recognition system. The pronunciation rules and the phone set were validated by test cases. The Arabic speech recognition system achieves word error rate of %11.71 for fully diacritized transcription of about 1.1 hours of Arabic broadcast news.

Abstract Sphinx-4 is a flexible, modular and pluggable framework to help foster new innovations in the core research of hidden Markov model (HMM) speech recognition systems. The design of Sphinx-4 is based on patterns that have emerged from the design of past systems as well as new requirements based on areas that researchers currently want to explore.

Abstract Statistical language modeling remains a challenging task, in particular for morphologically rich languages. Recently, new approaches based on factored language models have been developed to address this problem. These models provide principled ways of including additional conditioning variables other than the preceding words, such as morphological or syntactic features. However, the number of possible choices for model parameters creates a large space of models that cannot be searched exhaustively.