Large vocabulary continuous speech recognition for Urdu

Sarfraz, Huda; Hussain, Sarmad; Bokhari, Riffat; Raza, Agha Ali; Ullah, Inam; Sarfraz, Zahid; Pervez, Sophia; Mustafa, Asad; Javed, Iqra; Parveen, Rahila

doi:10.1145/1943628.1943629

Cited by 19 publications

(5 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They represent each word as a single phoneme which means that the system performance may suffer degradation for longer words. More recent work for continuous speech Urdu ASR is by Huda et al [12] using Sphinx open source toolkit. We are not making direct comparisons to Huda's work [12] as the data scale as well as application scale differ than our work.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Speech Recognition of Urdu Digits with Optimal Classification Approach

Ali¹,

An²,

Iqbal³

2015

IJCA

View full text Add to dashboard Cite

Speech Recognition for Urdu language is an interesting and less developed task. This is primarily due to the fact that linguistic resources such as rich corpus are not available for Urdu. Yet, few attempts have been made for developing Urdu speech recognition frameworks using the traditional approaches such as Hidden Markov Models and Neural Networks. In this work, we investigate the use of three classification methods for Urdu speech recognition task. We extract the Mel Frequency Cepstral Coefficients, the delta and delta-delta features from the speech data and train the classifiers to perform Urdu speech recognition. We present the performance achieved by training a Support Vector Machine (SVM) classifier, a random forest (RF) classifier and a linear discriminant analysis classifier (LDA) for comparison with SVM. Consequently, the experimental results show that SVM gives better performance than RF and LDA classifiers on this particular task.

show abstract

Section: Related Workmentioning

confidence: 99%

“…More recent work for continuous speech Urdu ASR is by Huda et al [12] using Sphinx open source toolkit. We are not making direct comparisons to Huda's work [12] as the data scale as well as application scale differ than our work. Besides, the speech data used in their experimentation is limited to one particular accent only.…”

Section: Related Workmentioning

confidence: 99%

Automatic Speech Recognition of Urdu Digits with Optimal Classification Approach

Ali¹,

An²,

Iqbal³

2015

IJCA

View full text Add to dashboard Cite

show abstract

“…However, the results are based on single speaker speech recognition and extensive enhancements are required to transform the system into a multi-speaker system. (Sarfraz et al 2010a ; 2010b ) has also used CMU Sphinx Toolkit towards Large Vocabulary speech recognition of Urdu. The goal was to cover the everyday speech; however, the variety in Urdu accents has not been covered as the target speech is mostly limited to suburban accent spoken in offices and homes.…”

Section: Related Workmentioning

confidence: 99%

DWT features performance analysis for automatic speech recognition of Urdu

et al. 2014

View full text Add to dashboard Cite

This paper presents the work on Automatic Speech Recognition of Urdu language, using a comparative analysis for Discrete Wavelets Transform (DWT) based features and Mel Frequency Cepstral Coefficients (MFCC). These features have been extracted for one hundred isolated words of Urdu, each word uttered by ten different speakers. The words have been selected from the most frequently used words of Urdu. A variety of age and dialect has been covered by using a balanced corpus approach. After extraction of features, the classification has been achieved by using Linear Discriminant Analysis. After the classification task, the confusion matrix obtained for the DWT features has been compared with the one obtained for Mel-Frequency Cepstral Coefficients based speech recognition. The framework has been trained and tested for speech data recorded under controlled environments. The experimental results are useful in determination of the optimum features for speech recognition task.

show abstract

“…Urdu continuous speech recognition system [32] was developed using spontaneous speech corpus collected from 82 speakers [25]. Speech corpus was recorded over telephone and microphone channels.…”

Section: Introductionmentioning

confidence: 99%

Improving Large Vocabulary Urdu Speech Recognition System Using Deep Neural Networks

et al. 2019

Self Cite

View full text Add to dashboard Cite

Development of Large Vocabulary Continuous Speech Recognition (LVCSR) system is a cumbersome task, especially for low resource languages. Urdu is the national language and lingua franca of Pakistan, with 100 million speakers worldwide. Due to resource scarcity, limited work has been done in the domain of Urdu speech recognition. In this paper, collection of Urdu speech corpus and development of Urdu speech recognition system is presented. Urdu LVCSR is developed using 300 hours of read speech data with a vocabulary size of 199K words. Microphone speech is recorded from 1671 Urdu and Punjabi speakers in both indoor and outdoor environments. Different acoustic modeling techniques such as Gaussian Mixture Models based Hidden Markov Models (GMM-HMM), Time Delay Neural Networks (TDNN), Long-Short Term Memory (LSTM) and Bidirectional Long-Short Term Memory (BLSTM) networks are investigated. Cross entropy and Lattice Free Maximum Mutual Information (LF-MMI) objective functions are employed during acoustic modeling. In addition, Recurrent Neural Network Language Model (RNNLM) is also being used for re-scoring. Developed speech recognition system has been evaluated on 9.5 hours of collected test data and a minimum Word Error Rate (%WER) of 13.50% is achieved.

show abstract

Large vocabulary continuous speech recognition for Urdu

Cited by 19 publications

References 8 publications

Automatic Speech Recognition of Urdu Digits with Optimal Classification Approach

Automatic Speech Recognition of Urdu Digits with Optimal Classification Approach

DWT features performance analysis for automatic speech recognition of Urdu

Improving Large Vocabulary Urdu Speech Recognition System Using Deep Neural Networks

Contact Info

Product

Resources

About