Publications/Speech Processing, Recognition & Audio Security
Conference · 2010
Large Vocabulary Continuous Speech Recognition for Urdu
Huda Sarfraz, Sarmad Hussain, Riffat Bokhari, Agha Ali Raza, Inam Ullah, Zahid Sarfraz, Sophia Pervez, Asad Mustafa, Iqra Javed, Rahila Parveen
Frontiers of Information Technology (FIT) 2010, Islamabad, Pakistan
Summary
This paper describes building the acoustic and language models for robust Urdu speech recognition with the open-source CMU Sphinx toolkit. Models were built incrementally, adding the speech of up to two speakers at a time: one model from 40 female speakers, one from 41 male speakers, and one combining all 81. We report recognition results and discuss approaches for improving them.
Cite this paper
Huda Sarfraz, Sarmad Hussain, Riffat Bokhari, Agha Ali Raza, Inam Ullah, Zahid Sarfraz, Sophia Pervez, Asad Mustafa, Iqra Javed, Rahila Parveen. Large Vocabulary Continuous Speech Recognition for Urdu. Frontiers of Information Technology (FIT) 2010, Islamabad, Pakistan.
@inproceedings{sarfraz2010large,
author = {Huda Sarfraz and Sarmad Hussain and Riffat Bokhari and Agha Ali Raza and Inam Ullah and Zahid Sarfraz and Sophia Pervez and Asad Mustafa and Iqra Javed and Rahila Parveen},
title = {{Large Vocabulary Continuous Speech Recognition for Urdu}},
booktitle = {Frontiers of Information Technology (FIT) 2010, Islamabad, Pakistan},
year = {2010},
url = {https://aghaaliraza.com/publications/2010--O-COCOSDA--Large_Vocabulary_Continuous_Speech_Recognition_for_Urdu.pdf}
}
- Dissecting ASR Failures in Low-Resource South Asian LanguagesInterspeech 2026, Sydney, Australia
- WER We Stand: Benchmarking Urdu ASR ModelsCOLING 2025, Abu Dhabi, UAE
- Conversations in the Wild: Data Collection, Automatic Generation and EvaluationComputer Speech & Language, vol. 89, 2025
- Deepfake Defense: Constructing and Evaluating a Specialized Urdu Deepfake Audio DatasetFindings of ACL 2024, Bangkok, Thailand
- Speech Emotion Recognition for the Urdu LanguageLanguage Resources and Evaluation (LRE), 2023