--- title: "Arabic Text-To-Speech (TTS) Data Preparation — Masri 2022" description: "This paper offers practical guidance on building recorded data for Arabic text-to-speech." authors: ["Hala Al Masri", "Muhy Eddin Za'ter"] year: 2022 venue: "arXiv:2204.03255" source: "https://arxiv.org/abs/2204.03255" doi: "10.48550/arxiv.2204.03255" arxiv: "2204.03255" licence: "CC BY 4.0 (arXiv)" licence_url: "http://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/masri-2022-arabic-text-speech-tts-data/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/masri-2022-arabic-text-speech-tts-data/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2022](https://md.research.elamur.ai/free/en/papers/2022/) / Arabic Text-To-Speech (TTS) Data Preparation # Arabic Text-To-Speech (TTS) Data Preparation **Authors:** Hala Al Masri, [Muhy Eddin Za'ter](https://md.research.elamur.ai/free/en/authors/muhy-eddin-za-ter/) · **Year:** 2022 · **Venue:** arXiv:2204.03255 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://arxiv.org/abs/2204.03255) · [PDF](https://arxiv.org/pdf/2204.03255) · [DOI 10.48550/arxiv.2204.03255](https://doi.org/10.48550/arxiv.2204.03255) · [arXiv 2204.03255](https://arxiv.org/abs/2204.03255) · [OpenAlex](https://openalex.org/W4226290868) **Citations:** 0 citations (openalex-snapshot · 2026-10-14) · 1 in our library, of the 140884 papers whose bibliographies we read ## Summary This paper offers practical guidance on building recorded data for Arabic text-to-speech. The authors argue that Arabic lags behind other languages in TTS and that the quality of the human-recorded ground truth largely decides how natural and intelligible the resulting voice will be. They combine a linguistic view with a studio-production view. They first outline a TTS pipeline aimed at banking and finance: text normalization, diacritization, spell checking, utterance segmentation and audio generation. They then describe the data work. - **Selecting the corpus:** open-domain or closed-domain text, phonetically balanced, with controlled utterance length. - **Diacritizing the text:** partial or full, to remove lexical ambiguity. - **Directing the recording sessions:** voice-actor selection, tempo, session length, accent, intonation, pauses and noise. Finally, they propose ways for listeners to judge speech quality: naturalness, degraded MOS, intelligibility and comprehensibility tests, plus a checklist of perceptual factors. The paper reports no experiments or released dataset. Its value lies in the specifications it gives to voice coaches, actors and annotators working on a low-resource, diacritic-dependent language. ## Key points - The proposed pipeline: normalization turns numbers, dates, abbreviations and foreign words into Arabic text, then diacritization, an Arabic spell checker, utterance segmentation, and finally audio generation. The target domain is finance and banking. - The data should be phonetically balanced and balanced in length. Phrases are capped at about 12 units, roughly what a speaker can say without pausing. - The paper tabulates eight Arabic diacritics with IPA and Buckwalter equivalents. Partial diacritization is presented as a cheaper way to resolve ambiguity, for example the many readings of an undiacritized word meaning "study/teach/lesson". - Recording rules include: - at most about 2 hours of recording per day; - a 10-minute break every hour; - returning to reference sentences after every 50 utterances; - re-recording instead of technical cuts; - pausing only at punctuation, with short vowels dropped at pauses; - keeping the delivery natural rather than acted. - Evaluation uses four tests: naturalness, degraded MOS with randomized natural and synthetic clips, intelligibility (MOS and WER from listener transcriptions), and comprehensibility (questions scored 0/1). ## Interesting ideas - Morphological analyzers can use whatever diacritics are present as hard filters, discarding any analysis that contradicts them, so sparse annotation still prunes ambiguity cheaply. — For Mova (tokens): Partial cues such as a known lemma, a capitalization hint or a silver POS can act as constraints that eliminate candidate analyses in the tagger, rather than as soft features. - Speakers periodically go back to a fixed set of reference sentences so that tempo and style do not drift over a long session. — For Mova (learn): A small frozen anchor set can be re-scored at regular intervals during training and self-play to catch drift before it contaminates silver data. - Defects are fixed by redoing the whole recording, never by splicing fragments, because patches break consistency. — For Mova (back): When generated text from the text → graph → text compressor has a local error, regenerating the whole sentence from the graph should yield more coherent output than patching words in place. - Intelligibility (writing down what was heard, regardless of meaning) is tested separately from comprehensibility (answering questions about the content). — For Mova (back): The compressor can be scored twice, once on surface reconstruction and once on whether Mova's readers answer the same questions on the regenerated text as on the original. - Evaluators rate a shuffled mix of real human recordings and synthetic ones, so the human items act as a built-in reference and bias check. — For Mova (learn): Gold human trees and frames can be hidden among model outputs when LLM teachers or witness parsers judge quality, serving as negative and positive controls for the judges themselves. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Linguistics & languages](https://md.research.elamur.ai/free/en/shelves/linguistics/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @misc{masri2022arabic, title = {{Arabic Text-To-Speech (TTS) Data Preparation}}, author = {Masri, Hala Al and Za'ter, Muhy Eddin}, year = {2022}, doi = {10.48550/arxiv.2204.03255}, eprint = {2204.03255}, archivePrefix = {arXiv}, url = {https://doi.org/10.48550/arxiv.2204.03255} } ``` - [← previous Arabic Fake News Detection Based on Deep Contextualized Embedding Models](https://md.research.elamur.ai/free/en/papers/nassif-2022-arabic-fake-news-detection-based/) - [next → Arabic Word-level Readability Visualization for Assisted Text Simplification](https://md.research.elamur.ai/free/en/papers/hazim-2022-arabic-word-level-readability-visualization/) - [↑ back to papers of 2022](https://md.research.elamur.ai/free/en/papers/2022/) ## References 14 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Paul Taylor. Text-to-speech synthesis. Cambridge university press, 2009. [find](https://scholar.google.com/scholar?q=Paul+Taylor%2E+Text%2Dto%2Dspeech+synthesis%2E+Cambridge+university+press%2C+2009%2E) 2. Andrew J Hunt and Alan W Black. Unit selection in a concatenative speech synthesis system using a large speech database. In 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings, volume 1, pages 373–376. IEEE, 1996. [find](https://scholar.google.com/scholar?q=Andrew+J+Hunt+and+Alan+W+Black%2E+Unit+selection+in+a+concatenative+speech+synthesis+system+using+a+large+speech+database%2E+In+1996+IEEE+International+Conference+on+Acoustics%2C+Speech%2C+and+Signal+Processi) 3. Alan W Black and Paul A Taylor. Automatically clustering similar units for unit selection in speech synthesis. 1997. [find](https://scholar.google.com/scholar?q=Alan+W+Black+and+Paul+A+Taylor%2E+Automatically+clustering+similar+units+for+unit+selection+in+speech+synthesis%2E+1997%2E) 4. Keiichi Tokuda, Takayoshi Yoshimura, Takashi Masuko, Takao Kobayashi, and Tadashi Kitamura. Speech parameter generation algorithms for hmm-based speech synthesis. In 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No. 00CH37100), volume 3, pages 1315–1318. IEEE, 2000. [find](https://scholar.google.com/scholar?q=Keiichi+Tokuda%2C+Takayoshi+Yoshimura%2C+Takashi+Masuko%2C+Takao+Kobayashi%2C+and+Tadashi+Kitamura%2E+Speech+parameter+generation+algorithms+for+hmm%2Dbased+speech+synthesis%2E+In+2000+IEEE+International+Conference) 5. Keiichi Tokuda, Yoshihiko Nankaku, Tomoki Toda, Heiga Zen, Junichi Yamagishi, and Keiichiro Oura. Speech synthesis based on hidden markov models. Proceedings of the IEEE, 101(5):1234–1252, 2013. [find](https://scholar.google.com/scholar?q=Keiichi+Tokuda%2C+Yoshihiko+Nankaku%2C+Tomoki+Toda%2C+Heiga+Zen%2C+Junichi+Yamagishi%2C+and+Keiichiro+Oura%2E+Speech+synthesis+based+on+hidden+markov+models%2E+Proceedings+of+the+IEEE%2C+101%285%29%3A1234%E2%80%931252%2C+2013%2E) 6. Heiga Zen, Keiichi Tokuda, and Alan W Black. Statistical parametric speech synthesis. speech communication, 51(11):1039–1064, 2009. [find](https://scholar.google.com/scholar?q=Heiga+Zen%2C+Keiichi+Tokuda%2C+and+Alan+W+Black%2E+Statistical+parametric+speech+synthesis%2E+speech+communication%2C+51%2811%29%3A1039%E2%80%931064%2C+2009%2E) 7. Heiga Ze, Andrew Senior, and Mike Schuster. Statistical parametric speech synthesis using deep neural networks. In 2013 ieee international conference on acoustics, speech and signal processing, pages 7962–7966. IEEE, 2013. [find](https://scholar.google.com/scholar?q=Heiga+Ze%2C+Andrew+Senior%2C+and+Mike+Schuster%2E+Statistical+parametric+speech+synthesis+using+deep+neural+networks%2E+In+2013+ieee+international+conference+on+acoustics%2C+speech+and+signal+processing%2C+pages+) 8. Aaron van den Oord, Sander Dieleman, Heiga Zen, Karen Simonyan, Oriol Vinyals, Alex Graves, Nal Kalchbrenner, Andrew Senior, and Koray Kavukcuoglu. Wavenet: A generative model for raw audio. arXiv preprint arXiv:1609.03499, 2016. [↗](https://arxiv.org/abs/1609.03499) [our card](https://md.research.elamur.ai/free/en/papers/oord-2016-wavenet-generative-model-raw-audio/) 9. Yuxuan Wang, RJ Skerry-Ryan, Daisy Stanton, Yonghui Wu, Ron J Weiss, Navdeep Jaitly, Zongheng Yang, Ying Xiao, Zhifeng Chen, Samy Bengio, et al. Tacotron: Towards end-to-end speech synthesis. arXiv preprint arXiv:1703.10135, 2017. [↗](https://arxiv.org/abs/1703.10135) 10. Akira Tamamori, Tomoki Hayashi, Kazuhiro Kobayashi, Kazuya Takeda, and Tomoki Toda. Speaker-dependent wavenet vocoder. In Interspeech, volume 2017, pages 1118–1122, 2017. [find](https://scholar.google.com/scholar?q=Akira+Tamamori%2C+Tomoki+Hayashi%2C+Kazuhiro+Kobayashi%2C+Kazuya+Takeda%2C+and+Tomoki+Toda%2E+Speaker%2Ddependent+wavenet+vocoder%2E+In+Interspeech%2C+volume+2017%2C+pages+1118%E2%80%931122%2C+2017%2E) 11. Jonathan Shen, Ruoming Pang, Ron J Weiss, Mike Schuster, Navdeep Jaitly, Zongheng Yang, Zhifeng Chen, Yu Zhang, Yuxuan Wang, Rj Skerrv-Ryan, et al. Natural tts synthesis by conditioning wavenet on mel spectrogram predictions. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 4779–4783. IEEE, 2018. [find](https://scholar.google.com/scholar?q=Jonathan+Shen%2C+Ruoming+Pang%2C+Ron+J+Weiss%2C+Mike+Schuster%2C+Navdeep+Jaitly%2C+Zongheng+Yang%2C+Zhifeng+Chen%2C+Yu+Zhang%2C+Yuxuan+Wang%2C+Rj+Skerrv%2DRyan%2C+et+al%2E+Natural+tts+synthesis+by+conditioning+wavenet+on+mel) 12. Taha Zerrouki, Mohammed M Abu Shquier, Amar Balla, Nabila Bousbia, Imededdine Sakraoui, and Fateh Boudardara. Adapting espeak to arabic language: Converting arabic text to speech language using espeak. International Journal of Reasoning-Based Intelligent Systems, 11(1):76–89, 2019. [find](https://scholar.google.com/scholar?q=Taha+Zerrouki%2C+Mohammed+M+Abu+Shquier%2C+Amar+Balla%2C+Nabila+Bousbia%2C+Imededdine+Sakraoui%2C+and+Fateh+Boudardara%2E+Adapting+espeak+to+arabic+language%3A+Converting+arabic+text+to+speech+language+using+espeak) 13. Fady K Fahmy, Mahmoud I Khalil, and Hazem M Abbas. A transfer learning end-to-end arabic text-to-speech (tts) deep architecture. In IAPR Workshop on Artificial Neural Networks in Pattern Recognition, pages 266–277. Springer, 2020. [find](https://scholar.google.com/scholar?q=Fady+K+Fahmy%2C+Mahmoud+I+Khalil%2C+and+Hazem+M+Abbas%2E+A+transfer+learning+end%2Dto%2Dend+arabic+text%2Dto%2Dspeech+%28tts%29+deep+architecture%2E+In+IAPR+Workshop+on+Artificial+Neural+Networks+in+Pattern+Recognition%2C+) 14. MOHAMED BEN MERIEM. Phonetic transcription of arabic. International Journal of Natural Language Processing, 1(1), 2020. [find](https://scholar.google.com/scholar?q=MOHAMED+BEN+MERIEM%2E+Phonetic+transcription+of+arabic%2E+International+Journal+of+Natural+Language+Processing%2C+1%281%29%2C+2020%2E)