Skip to main content
SpringerLink
Log in

Algerian Modern Colloquial Arabic Speech Corpus (AMCASC): regional accents recognition within complex socio-linguistic environments

  • Original Paper
  • Published:
Language Resources and Evaluation Aims and scope Submit manuscript

Abstract

The Algerian linguistic situation is very intricate due to the ethnic, geographical and colonial occupation influences which have lead to a complex sociolinguistic environment. As a result of the contact between different languages and accents, the Algerian speech community has acquired a distinctive sociolinguistic situation. In addition to the intra- and inter- lingual variations describing day-to-day linguistic behavior of the Algerian speakers, their speech is characterized by the presence of many linguistic phenomena such as bilingualism and code switching. The study of automatic regional accent recognition in such a type of environment is a new idea in the field of automatic languages, dialect and accent recognition especially that previous studies were conducted using monolingual evaluation data. The assessment of the effectiveness of GMM-UBM and i-vectors frameworks for accent recognition approaches through the use of the Algerian Modern Colloquial Arabic Speech Corpus (AMCASC), which is a linguistic resource collected for this purpose, shows that not only the recording conditions mismatch, channels mismatch, recordings length mismatch and the amplitude clipping which have a non-desirable effect on the effectiveness of these acoustic approaches but also language contact phenomena are other perturbation sources which should be taken into consideration especially in real life applications .

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

References

  • Abdelkader, B. L. (2011). The attitudes of Mzabi and Kabyle Minority Groups towards Code Switching and other Contact Phenomena in a Multilingual Sociolinguistic Environment. In The international Colloquium “Diversite et Innovation en Milieux Socioprofessionels”, November 16–18th at the MSH university of Clermont Ferrand, France.

  • Al-Ayyoub, M., Rihani, M. K., Dalgamoni, N. I., & Abdulla, N. A. (2014). Spoken Arabic dialects identification: The case of Egyptian and Jordanian dialects. In IEEE International conference on Information and communication systems (ICICS), (pp. 1–6). doi:10.1109/IACS.2014.6841970.

  • Alghamdi, M., Alhargan, F., Alkanhal, M., Alkhairy, A., Eldesouki, M., & Alenazi, A. (2008). Saudi Accented Arabic Voice Bank. Journal of King Saud University-Computer and Information Sciences, 20, 45–64.

    Article  Google Scholar 

  • Amino, K., & Osanai, T. (2014). Native versus non-native accent identification using Japanese spoken telephone numbers. Speech Communication, 56, 70–81. doi:10.1016/j.specom.2013.07.010.

    Article  Google Scholar 

  • Asmaa, B. (2012). The effects of trade activities on language variation at the Algerian/Moroccan border: Maghnia speech community. Tlemcen: University of Tlemcen.

    Google Scholar 

  • Bagui, H. (2014). Aspects of diglossic code switching situations: A sociolinguistic interpretation. European Journal of Research in Social Sciences, 2(4), 86–92.

    Google Scholar 

  • Bahari, M., Dehak, N., Burget, L., Ali, A. M., & Glass, J. (2014). Non-negative factor analysis of gaussian mixture model weight adaptation for language and dialect recognition. IEEE/ACM transactions on audio, speech, and language processing, 22(7), 1117–1129.

  • Bahari, M. H., Saeidi, R., & Van Leeuwen, D. (2013). Accent recognition using i-vector, gaussian mean supervector and gaussian posterior probability supervector for spontaneous telephone speech. In IEEE International conference on Acoustics, speech and signal processing (ICASSP) (pp. 7344–7348). doi:10.1109/ICASSP.2013.6639089.

  • Behravan, H., Hautamäki, V., & Kinnunen, T. (2013). Foreign accent detection from spoken finnish using i-vectors. In The 14th Annual Conference of the International Speech Communication Association (INTERSPEECH) (pp. 79–83). Lyon, France.

  • Behravan, H., Hautamäki, V., & Kinnunen, T. (2015). Factors affecting i-vector based foreign accent recognition: A case study in spoken Finnish. Speech Communication, 66, 118–129.

    Article  Google Scholar 

  • Biadsy, F. (2011). Automatic dialect and accent recognition and its application to speech recognition. New York: Columbia University.

    Google Scholar 

  • Biadsy, F., Hirschberg, J., & Habash, N. (2009). Spoken Arabic dialect identification using phonotactic modeling. In Proceedings of the EACL 2009 Workshop on Computational Approaches to Semitic Languages (pp. 53–61). Association for Computational Linguistics.

  • Bielefeld, B. (1994). Language identification using shifted delta cepstrum. Fourteenth Annual Speech Research Symposium.

  • Burget, L., Plchot, O., Cumani, S., Glembek, O., Matejka, P., & Brummer, N. (2011). Discriminatively trained probabilistic linear discriminant analysis for speaker verification. In IEEE International Conference on Acoustics, speech and signal processing (ICASSP) (pp. 4832–4835). doi:10.1109/ICASSP.2011.5947437.

  • Chen, N. F., Shen, W., & Campbell, J. P. (2010). A linguistically-informative approach to dialect recognition using dialect-discriminating context-dependent phonetic models. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing—Proceedings (pp. 5014–5017). doi:10.1109/ICASSP.2010.5495068.

  • Chen, N. F., Shen, W., Campbell, J. P., & Schwartz, R. (2009). Large-scale analysis of formant frequency estimation variability in conversational telephone speech. In The 10th Annual Conference of the International Speech Communication Association (INTERSPEECH) (pp. 2203–2206). Brighton, UK.

  • Chen, N. F., Shen, W., Campbell, J. P., & Torres-Carrasquillo, P. A. (2011). Informative dialect recognition using context-dependent pronunciation modeling. In IEEE International Conference on Acoustics, speech and signal processing (ICASSP) (pp. 4396–4399). doi:10.1109/ICASSP.2011.5947328.

  • Chen, N. F., Tam, S. W., Shen, W., & Campbell, J. P. (2014). Characterizing phonetic transformations and acoustic differences across english dialects. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 22(1), 110–124.

    Article  Google Scholar 

  • CNN. (2014). Voice words may provide key clues about James Foley’s Killer. Retrieved February 23, 2014, from http://edition.cnn.com/2014/08/22/world/europe/british-jihadi-hunt/.

  • Cumani, S., & Laface, P. (2012). Analysis of large-scale SVM training algorithms for language and speaker recognition. IEEE Transactions on Audio, Speech and Language Processing, 20(5), 1585–1596.

    Article  Google Scholar 

  • De Mareüil, P. B., & Bardiaux, A. (2011). Perception of French, Belgian and Swiss Accents by French and Belgian Listeners. In The 4th Workshop on Experimental Linguistics (ExLing) of the International Speech Communication Association (ISCA) (p. 47). Paris, France.

  • Dehak, N., Kenny, P., Dehak, R., Dumouchel, P., & Ouellet, P. (2011a). Front-end factor analysis for speaker verification. Audio, speech, and language processing. IEEE Transactions on Audio, Speech and Language Processing, 19(4), 788–798.

    Article  Google Scholar 

  • Dehak, N., Torres-Carrasquillo, P. A., Reynolds, D. A., & Dehak, R. (2011b). Language recognition via i-vectors and dimensionality reduction. In The 12th Annual Conference of the International Speech Communication Association (INTERSPEECH) (pp. 857–860). Florence, Italy.

  • DeMarco, A., & Cox, S. J. (2012). Iterative Classification of Regional British Accents in i-vector space. In Symposium on Machine MLSLP Learning in Speech and Language Processing (MLSLP) (pp. 1–4). Portland, Oregon.

  • Dendane, D. R. Z. (2013a). Linguistic variation and the age variable in an urban context: Tlemcen Speech Community. International Refereed Multidisciplinary Journal of Contemporary Research (IRMJCR), 1(03), 37–45.

    Google Scholar 

  • Dendane, Z. (2013b). The stigmatisation of the glottal stop in Tlemcen Speech Community: An indicator of dialect shift. International Journal of Linguistics and Literature (IJLL), 02(03), 1–10.

    Article  Google Scholar 

  • ELRA. (2013). Catalogue of language resources. Retrieved February 23, 2015, from http://catalog.elra.info/.

  • Feng, Z. (2011). A corpus for forensic identification of dialect speaker in some area. In 2011 IEEE 3rd International Conference on Communication Software and Networks (pp. 665–669). doi:10.1109/ICCSN.2011.6014814.

  • Gholipour, A., Sedaaghi, M. H., & Shamsi, M. (2012). The contribution of prosody to the identification of Persian Regional Accents. In 2012 IEEE Symposium on Industrial Electronics and Applications (pp. 346–350). IEEE. doi:10.1109/ISIEA.2012.6496658.

  • Hanani, A., Russell, M. J., & Carey, M. J. (2013). Human and computer recognition of regional accents and ethnic groups from British English Speech. Computer Speech & Language, 27(1), 59–74.

    Article  Google Scholar 

  • Huang, R., & Hansen, J. H. L. (2007). Unsupervised discriminative training with application to dialect classification. IEEE Transactions on Audio, Speech and Language Processing, 15(8), 2444–2453.

    Article  Google Scholar 

  • Huang, R., Hansen, J. H. L., & Angkititrakul, P. (2007). Dialect/accent classification using unrestricted audio. IEEE Transactions on Audio, Speech and Language Processing, 15(2), 453–464. doi:10.1109/TASL.2006.881695.

    Article  Google Scholar 

  • Jalalvand, S., Akbari, A., & Nasersharif, B. (2012). A classifier combination approach for Farsi accents recognition. In The 20th Iranian Conference on Electrical Engineering (ICEE) (pp. 716–720). doi:10.1109/IranianCEE.2012.6292447.

  • Kaouache, S. (2008). Dialect Stigma and Prejudice: The case of the Jijel dialect. Constantine: Mentouri University.

    Google Scholar 

  • Kenny, P., Boulianne, G., Ouellet, P., & Dumouchel, P. (2007). Joint factor analysis versus eigenchannels in speaker recognition. IEEE Transactions on Audio, Speech and Language Processing, 15(4), 1435–1447. doi:10.1109/TASL.2006.881693.

    Article  Google Scholar 

  • Khadidja, A. H. (2013). Language maintenance and language shift among Kabyle speakers in Arabic speaking communities: The case of Oran. Oran: University of Oran.

    Google Scholar 

  • Kockmann, M., & Ferrer, L. (2010). Prosodic speaker verification using subspace multinomial models with intersession compensation. In 11th Annual Conference of the International Speech Communication Association (INTERSPEECH) (pp. 1061–1064). Makuhari, Chiba, Japan.

  • Ladefoged, P. (2005). Vowels and consonants. New Jersey: Blackwell Publishing.

    Google Scholar 

  • Lazaridis, A., Khoury, E., Goldman, J.-P., Avanzi, M., Marcel, S., & Garner, P. N. (2014). Swiss French Regional Accent Identification. In The Speaker and Language Recognition Workshop (Odyssey) (pp. 106–111). Joensuu, Finland.

  • LDC. (1992). Linguistic data consortium. Retrieved February 23, 2015, from https://www.ldc.upenn.edu/.

  • Lei, Y., & Hansen, J. H. (2009). Factor analysis-based information integration for Arabic dialect identification. In IEEE International Conference on Acoustics, speech and signal processing (ICASSP) (pp. 4337–4340). doi:10.1109/ICASSP.2009.4960589.

  • Li, H., Ma, B, & Lee. K. A. (2013). Spoken language recognition : From fundamentals to practice. Proceedings of the IEEE, 101, 1136–1159. doi:10.1109/ICASSP.2009.4960589

  • Lincoln, M., Cox, S., & Ringland, S. (1998). A comparison of two unsupervised approaches to accent identification. In The 5th International Conference on Spoken Language Processing (ICSLP). Sydney, Australia.

  • Ma, B., Li, H., & Tong, R. (2007). Spoken language recognition using ensemble classifiers. IEEE Transactions on Audio, Speech and Language Processing, 15(7), 2053–2062.

    Article  Google Scholar 

  • Ma, B., Zhu, D., & Tong, R. (2006). Chinese dialect identification using tone features based on pitch flux. In IEEE International Conference on Acoustics Speed and Signal Processing Proceedings (ICASSP), (Vol. 1, pp. I–1029–I–1032). doi:10.1109/ICASSP.2006.1660199.

  • Malhotra, K., & Khosla, A. (2008). Automatic Identification of Gender & Accent in Spoken Hindi Utterances with Regional Indian Accents. In Spoken Language Technology Workshop, 2008. SLT 2008, (pp. 309–312). IEEE.

  • Meriem, B. M. (2014). A Sociolinguistic profile of French in Algeria: The case of Tlemcen Speech Community. Tlemcen: University of Tlemcen.

    Google Scholar 

  • Navratil, J. (2001). Spoken language recognition- A step toward multilinguality in speech processing. IEEE Transactions on Speech and Audio Processing, 9(6), 678–685.

    Article  Google Scholar 

  • Purnell, T., Idsardi, W., & Baugh, J. (1999). Perceptual and phonetic experiments on American English dialect identification. Journal of Language and Social Psychology, 18(1), 10–30. doi:10.1177/0261927X99018001002.

    Article  Google Scholar 

  • Rao, K. S., & Koolagudi, S. G. (2011). Identification of Hindi dialects and emotions using spectral and prosodic features of speech. IJSCI: International Journal of Systemics, Cybernetics and Informatics, 9(4), 24–33.

    Google Scholar 

  • Reynolds, D. A., Quatieri, T. F., & Dunn, R. B. (2000). Speaker verification using adapted gaussian mixture models. Digital Signal Processing, 10(2000), 19–41. doi:10.1006/dspr.1999.0361.

    Article  Google Scholar 

  • Selouani, S. A., & Boudraa, M. (2010). Algerian Arabic speech database (ALGASD): corpus design and automatic speech recognition application. Arabian Journal for Science and Engineering, 35(2C), 158.

    Google Scholar 

  • Singer, E., Torres-Carrasquillo, P., Reynolds, D., McCree, A., Richardson, F., Dehak, N., et al. (2012). In The MITLL NIST LRE 2011 language recognition system. The Speaker and Language Recognition Workshop (Odyssey) (pp. 209–215). Singapore.

  • STC. (2015). Smart Logger II: Multi-channel call recording and monitoring system. Retrieved February 23, 2015, from http://speechpro.com/product/voice-recording/smartlogger2.

  • Tan, B., Li, P., & Foresta, R. (2010). An automatic non-native speaker recognition system. In 2010 IEEE International Conference on Technologies for Homeland Security (HST) (pp. 77–83). doi:10.1109/THS.2010.5655088.

  • Tsai, W.-H., & Chang, W.-W. (2002). Discriminative training of gaussian mixture bigram models with application to Chinese dialect identification. Speech Communication, 36(3), 317–326.

    Article  Google Scholar 

  • Van der Vloed, D., Bouten, J., & van Leeuwen, D. A. (2014). NFI-FRITS: A forensic speaker recognition database and some first experiments. In Proceedings of Odyssey: The Speaker and Language Recognition Workshop, Joensuu, Finland.

  • Veaux, C., Yamagishi, J., & King, S. (2013). The Voice Bank Corpus: Design, collection and data analysis of a large regional accent speech database. In 2013 International Conference Oriental COCOSDA held jointly with 2013 Conference on Asian Spoken Language Research and Evaluation (O-COCOSDA/CASLRE) (pp. 1–4). IEEE. doi:10.1109/ICSDA.2013.6709856.

  • Versteegh, K., & Eid, M. (2006). Encyclopedia of Arabic language and linguistics. Boston.

  • Vieru, B., de Mareüil, P. B., & Adda-Decker, M. (2011). Characterisation and identification of non-native French accents. Speech Communication, 53(3), 292–310. doi:10.1016/j.specom.2010.10.002.

    Article  Google Scholar 

  • Wang, H., Leung, C.-C., Lee, T., Ma, B., & Li, H. (2013). Shifted-delta MLP features for spoken language recognition. IEEE Signal Processing Letters, 20(1), 15–18. doi:10.1109/LSP.2012.2227312.

    Article  Google Scholar 

  • Woehrling, C., & Boula de Mareuil, P. (2006). Identification d’Accents Regionaux en Francais: Perception et Analyse. Revue Parole, 37, 55.

    Google Scholar 

  • Youcef, B. (2007). The Phonological Interlanguage of the Undergraduate Students at the Department of English, University of Mentouri, Constantine. Constantine: University of Mentouri.

    Google Scholar 

  • Zhang, X., & Demuynck, K. (2013). Rapid speaker adaptation in latent speaker space with non-negative matrix factorization. Speech Communication, 55(9), 893–908.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. LCPTS, FEI, USTHB, 16111, Bab Ezzouar, Algeria

    Mourad Djellab & Abderrahmane Amrouche

  2. Department of Computer Science and Digital Technologies, Northumbria University, Newcastle upon Tyne, NE1 8ST, UK

    Ahmed Bouridane

  3. CRD-GN, 16002, Bouchaoui, Algeria

    Noureddine Mehallegue

Authors
  1. Mourad Djellab
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Abderrahmane Amrouche
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ahmed Bouridane
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Noureddine Mehallegue
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Mourad Djellab or Ahmed Bouridane.

Appendices

Appendix 1: MCA linguistic features in some Algerian regions

The sociolinguistic situation in Algeria is characterized by a diversity and coexistence of different languages which are the Arabic and the Tamazight language in addition to the French language. The Arabic language has two forms: the Modern Standard Arabic (MSA or the classical Arabic) which is the official language used at schools, Media and the Algerian administration but it is not the daily language, and the Modern colloquial Arabic (MCA) which is an Arabic dialect known as the Algerian Arabic. This language is used in everyday activities and it is the mother tongue of the majority of Algerians. It varies considerably across the country (Selouani and Boudraa 2010).

The Tamazight is a language spoken by a large population in Algeria. The main Amazigh dialects in this country are: Kabyle which is spoken in the area of Djurdjura, Chaoui spoken in the mountains of the Aures (east of Algiers) and Mozabit spoken in the northern Sahara, in addition to the Tuareg spoken in the south and the Chleuh spoken in the area of Chenoua. Some of these dialects consist of several regional accents which are characterized by a wide variation mainly in phonetics and vocabulary. Despite this fact, the mutual understanding between different speakers is quite easy for people of the same dialect.

As a result of the contact between all these languages, there are significant local variations (in pronunciation, grammar, etc.) of spoken Arabic (MSA and MCA) in Algeria. However, before discussing differences in linguistic features observed in MCA across Algeria, it is worth noting that to the best of our knowledge, there is no published work on this specific problem. This situation makes the investigation of this issue a challenging task. Thus, our description in this section focuses on some specific linguistic features characterising the well-known Algerian accents. These features are generally related to the phonologically, morphologically and lexically language variations.

1.1 Tlemcen Arabic (TA)

One of the most known regional accents in Algeria is Tlemcen Arabic (TA) which is used in the extreme North West borders of Algeria with Morocco. The variety of Arabic used in this region (TA), has long distinguished itself from other Algerian Arabic accents by a number of linguistic features (Dendane 2013b). It characterizes itself by an old urban phonological trait, not used anywhere else in the country, is the realisation of /q/, the Classical Arabic (CA) qāf, as a glottal stop [ʔ], as in [ʔæ:l], ‘He said’, for CA /qaal/ (Dendane 2013b). Other common phonological features which are related to the speech community of this region are the alveolar [z] which is articulated [ʒ], for example, the word “marriage” which is: /zawa:ʒ/ in CA, is pronounced in TA: [zwæʒ] or [ʒwæʒ]. The TA differs also from other Algerian accents in some morphological features. The internal structure of words and the alternation through the addition of prefixes and suffixes are some characteristic used in TA accent. As examples, we mention the addition of the prefix (ka) to the verb “I like”, or [nabɤɪ] in CA, which is pronounced [ka + nabɤɪ] in TA. Also, the use of the suffix morpheme [i] when addressing both males and females: “Where were you?” they pronounced it [fi:n # kunt] for male and [fi:n # kuntɪ] for female. The other linguistic particularity of TA is lexical, represented mainly in the daily use of Moroccan borrowings imported from the near-by Moroccan cities (Asmaa 2012).

1.2 Jijel Arabic (JA)

The second Algerian Arabic accents is Jijel Arabic (JA) accent, where the phoneme [q] is replaced by [k]; [ð] is pronounced as [d]; [dˤ] and [ðˤ] are pronounced as [tˤ]; and [θ] and [t] are pronounced as [ts]. JA accent is also different from other varieties in Algeria by its linguistic items (sound-patterns and constructions) which are typical to this accent and which are highly stigmatized and not easy to classify within the class of lexicons, or sound-patterns, or constructions (Kaouache 2008). One of these items are: the |h0a| which is equivalent to the determiner “a” in English, and the |ddi| which is used as a preposition exactly the same as the French preposition “de”. Negation and deletion are other characteristics of Jijel accent, such expressions as |hadak| and |hadik| (‘that one’) are often heard |dak| and |dik|, respectively in Jijel accent. The JA is also characterized by the absence of emphasis; speakers from Jijel are quickly categorized by the other Algerian native speakers, when using non-emphatic sounds. For example, the listener to JA speakers may hear |d0|, |d|, and |t0| for the same sound, and sometimes can understand their meaning only within the context (Kaouache 2008) (Table 8).

Table 8 Examples illustrating the particularity of Jijel Arabic
Full size table

Appendix 2: Experimental results using the i-vectors framework

The experimental results of the evaluation of the i-vectors approach using the data collected from three regional Algerian accents are reported in Table 9.

Table 9 Recognition rates obtained using i-vectors approach
Full size table

As shown in the previous table, the i-vectors technique does not provide significant improvements when compared with GMM-UBM. This may be due to various factors such as the development data which was used to train the total variability matrix (T), the ability of cepstral features such as shifted delta cepstrum to reflect the difference existing between the investigated regional accents, as well as the dimensionality reduction process effect (Behravan et al. 2015). Our results are in perfect agreement with those reported in (DeMarco and Cox 2012) on a British regional accent dataset in the sense that no gain was obtained with i-vectors if GMM-SVM is used as a reference point.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Djellab, M., Amrouche, A., Bouridane, A. et al. Algerian Modern Colloquial Arabic Speech Corpus (AMCASC): regional accents recognition within complex socio-linguistic environments. Lang Resources & Evaluation 51, 613–641 (2017). https://doi.org/10.1007/s10579-016-9347-6

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10579-016-9347-6

Keywords

Search

Navigation