Authors :
Jean-Christophe B. Wizamo; Selain K. Kasereka
Volume/Issue :
Volume 11 - 2026, Issue 11 - November
Google Scholar :
https://tinyurl.com/mw9phzz
DOI :
https://doi.org/10.38124/ijisrt/26nov245
Note : A published paper may take 4-5
working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and
ResearchGate.
Abstract :
French is the sole official language of the Democratic Republic of the Congo, yet much of the population communicates daily in Lingala, Swahili, Tshiluba or Kikongo. This paper presents a cascaded system that transcribes spoken French with Whisper and translates it with NLLB-200-distilled-600M, specialized by one Low-Rank Adaptation (LoRA) adapter per target language on a shared frozen base model.
Keywords :
Low-Resource Machine Translation; Speech Translation; Whisper; NLLB-200; LoRA; Lingala; Swahili; Tshiluba; Kikongo; Edge AI.
References :
- République Démocratique du Congo, Constitution de la République Démocratique du Congo. Kinshasa, Democratic Republic of the Congo: Journal Officiel de la RDC, Feb. 18, 2006, art. 1.
- R. E. Mandiya et al., “A transformer-based method for bidirectional French–Lingala machine translation in speech and text,” Appl. Sci., vol. 16, no. 7, art. no. 3399, 2026, doi: 10.3390/app16073399.
- W. Nekoto et al., “Participatory research for low-resourced machine translation: A case study in African languages,” in Findings of the Association for Computational Linguistics: EMNLP 2020, Online, 2020, pp. 2144–2160.
- A. Radford et al., “Robust speech recognition via large-scale weak supervision,” in Proc. 40th Int. Conf. Machine Learning (ICML), Honolulu, HI, USA, 2023, vol. 202, pp. 28492–28518.
- NLLB Team et al., “No language left behind: Scaling human-centered machine translation,” arXiv:2207.04672, 2022.
- Ž. Agić and I. Vulić, “JW300: A wide-coverage parallel corpus for low-resource languages,” in Proc. 57th Annu. Meeting of the Association for Computational Linguistics (ACL), Florence, Italy, 2019, pp. 3204–3210.
- Y. Mao, C. You, J. Zhang, K. Huang, and K. B. Letaief, “A survey on mobile edge computing: The communication perspective,” IEEE Commun. Surveys Tuts., vol. 19, no. 4, pp. 2322–2358, 2017.
- J. Kreutzer et al., “Quality at a glance: An audit of web-crawled multilingual datasets,” Trans. Assoc. Comput. Linguistics, vol. 10, pp. 50–72, 2022.
- E. J. Hu et al., “LoRA: Low-rank adaptation of large language models,” in Proc. 10th Int. Conf. Learning Representations (ICLR), Virtual, 2022.
- B. Jacob et al., “Quantization and training of neural networks for efficient integer-arithmetic-only inference,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, 2018, pp. 2704–2713.
- A. Gholami et al., “A survey of quantization methods for efficient neural network inference,” arXiv:2103.13630, 2021.
- A. Bérard, O. Pietquin, C. Servan, and L. Besacier, “Listen and translate: A proof of concept for end-to-end speech-to-text translation,” in Proc. NIPS Workshop on End-to-End Learning for Speech and Audio Processing, Barcelona, Spain, 2016.
- R. J. Weiss, J. Chorowski, N. Jaitly, Y. Wu, and Z. Chen, “Sequence-to-sequence models can directly translate foreign speech,” in Proc. Interspeech, Stockholm, Sweden, 2017, pp. 2625–2629.
- M. Sperber and M. Paulik, “Speech translation and the end-to-end promise: Taking stock of where we are,” in Proc. 58th Annu. Meeting of the Association for Computational Linguistics (ACL), Online, 2020, pp. 7409–7421.
- Seamless Communication et al., “SeamlessM4T: Massively multilingual and multimodal machine translation,” arXiv:2308.11596, 2023.
- V. Pratap et al., “Scaling speech technology to 1,000+ languages,” arXiv:2305.13516, 2023.
- M. Ma et al., “STACL: Simultaneous translation with implicit anticipation and controllable latency using prefix-to-prefix framework,” in Proc. 57th Annu. Meeting of the Association for Computational Linguistics (ACL), Florence, Italy, 2019, pp. 3025–3036.
- D. Macháček, R. Dabre, and O. Bojar, “Turning Whisper into real-time transcription system,” in Proc. 13th Int. Joint Conf. Natural Language Processing and 3rd Conf. Asia-Pacific Chapter of the ACL: System Demonstrations, Bali, Indonesia, 2023.
- I. Sutskever, O. Vinyals, and Q. V. Le, “Sequence to sequence learning with neural networks,” in Advances in Neural Information Processing Systems, vol. 27, 2014.
- D. Bahdanau, K. Cho, and Y. Bengio, “Neural machine translation by jointly learning to align and translate,” in Proc. 3rd Int. Conf. Learning Representations (ICLR), San Diego, CA, USA, 2015.
- A. Vaswani et al., “Attention is all you need,” in Advances in Neural Information Processing Systems, vol. 30, 2017, pp. 5998–6008.
- M. Johnson et al., “Google’s multilingual neural machine translation system: Enabling zero-shot translation,” Trans. Assoc. Comput. Linguistics, vol. 5, pp. 339–351, 2017.
- N. Arivazhagan et al., “Massively multilingual neural machine translation in the wild: Findings and challenges,” arXiv:1907.05019, 2019.
- A. Fan et al., “Beyond English-centric multilingual machine translation,” J. Mach. Learn. Res., vol. 22, pp. 1–48, 2021.
- M. Artetxe and H. Schwenk, “Massively multilingual sentence embeddings for zero-shot cross-lingual transfer and beyond,” Trans. Assoc. Comput. Linguistics, vol. 7, pp. 597–610, 2019.
- N. Goyal et al., “The Flores-101 evaluation benchmark for low-resource and multilingual machine translation,” Trans. Assoc. Comput. Linguistics, vol. 10, pp. 522–538, 2022.
- D. I. Adelani et al., “A few thousand translations go a long way! Leveraging pre-trained models for African news translation,” in Proc. 2022 Conf. North American Chapter of the ACL: Human Language Technologies (NAACL-HLT), Seattle, WA, USA, 2022, pp. 3053–3070.
- N. Houlsby et al., “Parameter-efficient transfer learning for NLP,” in Proc. 36th Int. Conf. Machine Learning (ICML), Long Beach, CA, USA, 2019, vol. 97, pp. 2790–2799.
- A. Bapna and O. Firat, “Simple, scalable adaptation for neural machine translation,” in Proc. 2019 Conf. Empirical Methods in Natural Language Processing and 9th Int. Joint Conf. Natural Language Processing (EMNLP-IJCNLP), Hong Kong, China, 2019, pp. 1538–1548.
- M. Nagel et al., “A white paper on neural network quantization,” arXiv:2106.08295, 2021.
- O. Zafrir, G. Boudoukh, P. Izsak, and M. Wasserblat, “Q8BERT: Quantized 8Bit BERT,” in Proc. 5th Workshop on Energy Efficient Machine Learning and Cognitive Computing (EMC2-NeurIPS), Vancouver, BC, Canada, 2019, pp. 36–39.
- Y. Bondarenko, M. Nagel, and T. Blankevoort, “Understanding and overcoming the challenges of efficient Transformer quantization,” in Proc. 2021 Conf. Empirical Methods in Natural Language Processing (EMNLP), Punta Cana, Dominican Republic, 2021, pp. 7947–7969.
- G. Hinton, O. Vinyals, and J. Dean, “Distilling the knowledge in a neural network,” arXiv:1503.02531, 2015.
- S. Gandhi, P. von Platen, and A. M. Rush, “Distil-Whisper: Robust knowledge distillation via large-scale pseudo labelling,” arXiv:2311.00430, 2023.
- ONNX Runtime Developers, “Quantize ONNX models,” ONNX Runtime documentation. [Online]. Available: https://onnxruntime.ai/docs/performance/model-optimizations/quantization.html (accessed Oct. 21, 2026).
- M. T. Mukungu, A. M. Mbayandjambe, T. Tashev, K. Kyamakya, and S. K. Kasereka, “BabaSpeech: A deep learning-based translation of sign language into Lingala text and speech for deaf-mute inclusivity,” in Proc. Int. Joint Conf. Artificial Intelligence, Singapore: Springer, 2025, pp. 91–105.
- A. M. Mbayandjambe, S. K. Kasereka, K. Kyamakya, and V. T. Ho, “Enhancing printed Lingala script recognition using deep learning techniques,” Procedia Comput. Sci., vol. 257, pp. 111–118, 2025.
- Silero Team, “Silero VAD: Pre-trained enterprise-grade voice activity detector,” GitHub repository, 2021. [Online]. Available: https://github.com/snakers4/silero-vad
- J. Tiedemann, “The Tatoeba translation challenge – Realistic data sets for low resource and multilingual MT,” in Proc. 5th Conf. Machine Translation (WMT), Online, 2020, pp. 1174–1182.
- A. Anastasopoulos et al., “TICO-19: The translation initiative for COvid-19,” in Proc. 1st Workshop on NLP for COVID-19 (Part 2) at EMNLP 2020, Online, 2020.
- J. Tiedemann, “Parallel data, tools and interfaces in OPUS,” in Proc. 8th Int. Conf. Language Resources and Evaluation (LREC), Istanbul, Turkey, 2012, pp. 2214–2218.
- T. Kudo and J. Richardson, “SentencePiece: A simple and language independent subword tokenizer and detokenizer for neural text processing,” in Proc. 2018 Conf. Empirical Methods in Natural Language Processing: System Demonstrations, Brussels, Belgium, 2018, pp. 66–71.
- R. Sennrich, B. Haddow, and A. Birch, “Neural machine translation of rare words with subword units,” in Proc. 54th Annu. Meeting of the Association for Computational Linguistics (ACL), Berlin, Germany, 2016, pp. 1715–1725.
- I. Loshchilov and F. Hutter, “Decoupled weight decay regularization,” in Proc. 7th Int. Conf. Learning Representations (ICLR), New Orleans, LA, USA, 2019.
- S. Migacz, “8-bit inference with TensorRT,” presented at the GPU Technology Conf., San Jose, CA, USA, May 2017.
- A. Abid et al., “Gradio: Hassle-free sharing and testing of ML models in the wild,” arXiv:1906.02569, 2019.
- K. Papineni, S. Roukos, T. Ward, and W.-J. Zhu, “BLEU: A method for automatic evaluation of machine translation,” in Proc. 40th Annu. Meeting of the Association for Computational Linguistics (ACL), Philadelphia, PA, USA, 2002, pp. 311–318.
- M. Popović, “chrF: Character n-gram F-score for automatic MT evaluation,” in Proc. 10th Workshop on Statistical Machine Translation (WMT), Lisbon, Portugal, 2015, pp. 392–395.
- M. Post, “A call for clarity in reporting BLEU scores,” in Proc. 3rd Conf. Machine Translation: Research Papers (WMT), Brussels, Belgium, 2018, pp. 186–191.
- T. Kocmi et al., “To ship or not to ship: An extensive evaluation of automatic metrics for machine translation,” in Proc. 6th Conf. Machine Translation (WMT), Online, 2021, pp. 478–494.
- T. Wolf et al., “Transformers: State-of-the-art natural language processing,” in Proc. 2020 Conf. Empirical Methods in Natural Language Processing: System Demonstrations, Online, 2020, pp. 38–45.
- S. Mangrulkar et al., “PEFT: State-of-the-art parameter-efficient fine-tuning methods,” GitHub repository, 2022. [Online]. Available: https://github.com/huggingface/peft.
French is the sole official language of the Democratic Republic of the Congo, yet much of the population communicates daily in Lingala, Swahili, Tshiluba or Kikongo. This paper presents a cascaded system that transcribes spoken French with Whisper and translates it with NLLB-200-distilled-600M, specialized by one Low-Rank Adaptation (LoRA) adapter per target language on a shared frozen base model.
Keywords :
Low-Resource Machine Translation; Speech Translation; Whisper; NLLB-200; LoRA; Lingala; Swahili; Tshiluba; Kikongo; Edge AI.