An enhanced automatic speech recognition system for Arabic

@inproceedings{Menacer2017AnEA,
  title={An enhanced automatic speech recognition system for Arabic},
  author={Mohamed Amine Menacer and Odile Mella and D. Fohr and Denis Jouvet and David Langlois and Kamel Sma{\"i}li},
  booktitle={WANLP@EACL},
  year={2017}
}
Automatic speech recognition for Arabic is a very challenging task. [] Key Method We develop an ASR system for MSA by using Kaldi toolkit. Several acoustic and language models are trained. We obtain a Word Error Rate (WER) of 14.42 for the baseline system and 12.2 relative improvement by rescoring the lattice and by rewriting the output with the right hamoza above or below Alif.

Figures and Tables from this paper

Construction of a database for speech recognition of isolated Arabic words
TLDR
The paper presents the significance of the ASR systems built in the past few years and introduces a new Arabic database for isolated word by defining a new concept of phonetic units: semi-syllable units.
Multi-Dialect Arabic Speech Recognition
  • Abbas Raza Ali
  • Computer Science
    2020 International Joint Conference on Neural Networks (IJCNN)
  • 2020
TLDR
The design and development of multi-dialect automatic speech recognition for Arabic with a 14% error rate is presented and the development of a framework to train an acoustic model achieving state-of-the-art performance is developed.
Concatenative Speech Recognition using Morphemes
TLDR
The paper shows that the approach used encompasses fundamentally different processes of word formation and thus is applicable to languages that exhibit concatenative word-formation processes.
Automatic Speech Recognition for Tunisian Dialect
TLDR
The first steps to build an automatic speech recognition system for Tunisian dialect are proposed, in this paper, using HMM-DNN system, which can give an impressive relative reduction in WER.
A Dataset for Speech Recognition to Support Arabic Phoneme Pronunciation
TLDR
An automatic speech recognition system which has the capacity to detect the incorrect phoneme pronunciation and can automatically support children to improve their pronunciation by directly asking children to pronounce a phoneme and the system can tell them if it is correct or not.
Effects of Language Modelling for Sepedi-English Code-Switched Speech in Automatic Speech Recognition System
TLDR
The Witten-Bell smoothing technique was found to perform better than the other three smoothing techniques for Sepedi-English CS data in Language Modelling and also in the authors' ASR.
Is statistical machine translation approach dead
TLDR
This article aims to describe some of powerful and advanced techniques proposed to improve the NMT system and to compare them with the conventional SMT approach on the task of Arabic-English machine translation.
Robustness of end-to-end Automatic Speech Recognition Models – A Case Study using Mozilla DeepSpeech
TLDR
It is argued that many performance numbers reported probably underestimate the expected error rate, and it is found that content overlap has the biggest impact, but other factors like gender also play a role.
A Fine-Grained Multilingual Analysis Based on the Appraisal Theory: Application to Arabic and English Videos
TLDR
A fine-grained approach inspired from the appraisal theory is used to analyze the content of the videos that concern the same topic and considers more detailed sentiments by covering additional attributes of opinions such as: Attitude, Graduation and Engagement.
...
1
2
...

References

SHOWING 1-10 OF 33 REFERENCES
Automatic Speech-to-Text Transcription in Arabic
TLDR
The initial research was oriented toward processing of broadcast news data in Modern Standard Arabic, and has been extended to address a larger variety of broadcast data, which results in the need to also be able to handle dialectal speech.
Morphological Decomposition for Arabic Broadcast News Transcription
TLDR
A novel approach for morphological decomposition in large vocabulary Arabic speech recognition achieved low out- of-vocabulary (OOV) rate as well as high recognition accuracy in a state-of-the-art Arabic broadcast news transcription system.
Investigating the use of morphological decomposition and diacritization for improving Arabic LVCSR
TLDR
This work is able to obtain about 3.7% relative reduction in word error rate (WER) with respect to a comparable non-diacritized full-words system running on the authors' test set.
On the use of morphological analysis for dialectal Arabic speech recognition
TLDR
A simple word decomposition algorithm is introduced which only requires a text corpus and a predefined list of affixes to create the lexicon for Iraqi Arabic ASR and results in about 10% relative improvement in word error rate (WER).
A complete KALDI recipe for building Arabic speech recognition systems
TLDR
A prototype broadcast news system using 200 hours GALE data that is publicly available through LDC and the first effort to share reproducible sizable training and testing results on MSA system is shared.
Morpheme-Based Language Modeling for Arabic Lvcsr
TLDR
This paper uses morpheme-based language modeling to improve the word error rate in Arabic, and proposes a simple constraining method to rid the decoding output of illegal morphe me sequences.
Morphological analysis and decomposition for Arabic speech-to-text systems
TLDR
A novel context-sensitive method for morpheme-to-word conversion is introduced and the performance of the MADA decomposed system was evaluated on an Arabic broadcast transcription task, with both the morphological decomposition and stem normalisation being found to be important.
Improved morphological decomposition for Arabic broadcast news transcription
We show the progress for Arabic speech recognition by incorporating contextual information into the process of morphological decomposition. The new approach achieves lower out-of-vocabulary and word
Improved Spelling Error Detection and Correction for Arabic
TLDR
This work semi-automatically develops a dictionary of 9.3 million fully inflected Arabic words using a morphological transducer and a large corpus and improves the error model and language model.
Arabic Spelling Correction using Supervised Learning
In this work, we address the problem of spelling correction in the Arabic language utilizing the new corpus provided by QALB (Qatar Arabic Language Bank) project which is an annotated corpus of
...
1
2
3
4
...