Elamur Research⌕
☰
home / papers / 2026 / FalAR: A Large-scale Speaker-Annotated European Portuguese Speech Corpus of Parliamentary Sessions

FalAR: A Large-scale Speaker-Annotated European Portuguese Speech Corpus of Parliamentary Sessions

Summary

Speech recognition for European Portuguese (EP) lags behind because large Portuguese speech collections are dominated by Brazilian Portuguese. Brazilian Portuguese has roughly 200 million speakers, while EP has about 11 million. Labeled EP data has stayed small, and the 425-hour CAMÕES training set is among the largest. The authors built FalAR from publicly released recordings and manual transcripts of the Portuguese parliament covering about two decades. They aligned the official transcripts with pseudo-transcriptions produced by a CAMÕES EP ASR model. The result is 5,800 hours of aligned audio, of which 4,850 hours are linked to 1,180 identified speakers with demographic and political metadata. The authors trained ASR models on subsets of increasing size and alignment error. Adding more data kept helping even though the alignments became noisier, and out-of-domain models approached in-domain baselines on the CAMÕES benchmark. Pre-training on FalAR before in-domain fine-tuning reduced WER by up to 14% relative to training from scratch. Unlike the multilingual EuroSpeech corpus, FalAR offers speaker labels and EP-specific alignment, which supports both ASR and speaker recognition for EP.

Key points

  • Corpus size: FalAR contains about 5,800 hours of transcript-aligned EP parliamentary speech spanning roughly 20 years. It is released at huggingface.co/datasets/inesc-id/FalAR.
  • Speaker annotation: 4,850 hours carry speaker identities for 1,180 unique speakers. Each speaker has metadata on age, gender, political affiliation and parliamentary role, which enables longitudinal studies of up to 20 years.
  • Alignment method: Official transcripts were aligned using pseudo-transcriptions from a state-of-the-art CAMÕES EP ASR model.
  • ASR results: Experiments weighed the benefit of using more training data against the cost of less precise alignments. Pre-training on FalAR gave up to 14% relative WER improvement over a domain-specific baseline.
  • Context: Earlier labeled EP resources were much smaller, including BD-PUBLICO (25 h), ALERT (74 h), SpeechDat EP (81 h) and the CAMÕES training data (425 h). FalAR complements EuroSpeech, which lacks speaker metadata.

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

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/2xhv97bm5dyd

@inproceedings{teixeira2026falar,
  title = {{FalAR: A Large-scale Speaker-Annotated European Portuguese Speech Corpus of Parliamentary Sessions}},
  author = {Teixeira, Francisco and Carvalho, Carlos and Julião, Mariana and Botelho, Catarina and Solera-Ureña, Rubén and Paulo, Sérgio and Rolland, Thomas and Peters, Ben and Trancoso, Isabel and Abad, Alberto},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/2xhv97bm5dyd},
  url = {https://doi.org/10.63317/2xhv97bm5dyd}
}

References

57 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Alberto Abad, Isabel Trancoso, Nelson Neto, and M. Céu Viana. 2009. Porting an european portuguese broadcast news recognition system to brazilian portuguese. In Interspeech 2009, pages 92–95. find
  2. José Aires, Aida Cardoso, Rui Pereira, and Amália Mendes. 2024. Compiling and exploring a Portuguese parliamentary corpus: ParlaMint-PT. In Proc. of the IV Workshop on Creating, Analysing, and Increasing Accessibility of Parliamentary Corpora (ParlaCLARIN)@ LREC-COLING 2024, pages 12–20. find
  3. Vladimir Fabregas Surigué de Alencar and Abraham Alcaim. 2008. LSF and LPC-derived features for large vocabulary distributed continuous speech recognition in Brazilian Portuguese. In Proc. 42nd Asilomar Conference on Signals, Systems and Computers, pages 1237–1241. find
  4. Paulo Almeida, Manuel Marques-Pita, and Joana Gonçalves-Sá. 2021. PTPARL-D: an annotated corpus of forty-four years of Portuguese parliamentary debates. Corpora, 16(3):337–348. find
  5. Mohamed Anwar, Bowen Shi, Vedanuj Goswami, Wei-Ning Hsu, Juan Pino, and Changhan Wang. 2023. MuAViC: A Multilingual Audio-Visual Corpus for Robust Speech Recognition and Robust Speech-to-Text Translation. In Proc. Interspeech, pages 4064–4068. find
  6. Rosana Ardila, Megan Branson, Kelly Davis, Michael Kohler, Josh Meyer, Michael Henretty, Reuben Morais, Lindsay Saunders, Francis Tyers, and Gregor Weber. 2020. Common Voice: A Massively-Multilingual Speech Corpus. In Proc. LREC, pages 4218–4222. find our card
  7. Hervé Bredin. 2023. pyannote.audio 2.1 speaker diarization pipeline: principle, benchmark, and recipe. In Proc. find
  8. Adriano Campinho. 2021. Automatic speech recognition for European Portuguese. Master’s thesis, Escola de Engenharia, Universidade do Minho, Braga, Portugal, July. Available at https://hdl.handle.net/1822/78249. ↗
  9. Arnaldo Candido Junior, Edresson Casanova, Anderson Soares, Frederico Santos de Oliveira, Lucas Oliveira, Ricardo Corso Fernandes Junior, Daniel Peixoto Pinto da Silva, Fernando Gorgulho Fayet, Bruno Baldissera Carlotto, Lucas Rafael Stefanel Gris, and Sandra M. Aluísio. 2023. CORAA ASR: a large cor… find
  10. Carlos Carvalho. 2021. TRIBUS: An end-to-end automatic speech recognition system for European Portuguese. Master’s thesis, Instituto Superior Técnico, Universidade de Lisboa, Lisbon, Portugal, January. Available at https://fenix.tecnico.ulisboa.pt/downloadFile/1126295043839127/ 81395-carlos-carvalh… ↗
  11. Carlos Carvalho and Alberto Abad. 2021. TRIBUS: An end-to-end automatic speech recognition system for European Portuguese. In Proc. Iber-SPEECH, pages 185–189. find
  12. Carlos Carvalho, Francisco Teixeira, Catarina Botelho, Anna Pompili, Rubén Solera-Ureña, Sérgio Paulo, Mariana Julião, Thomas Rolland, John Mendonça, Diogo Pereira, Isabel Trancoso, and Alberto Abad. 2025. CAMÕES: A comprehensive automatic speech recognition benchmark for European Portuguese. In Acc… find
  13. Centro de Estudos Humanísticos, Universidade do Minho. 2009. Perfil Sociolinguístico da Fala Bracarense. https://sites.google.com/site/projectofalabracarense/. Accessed: 2025-10-24. ↗
  14. William Chan, Daniel Park, Chris Lee, Yu Zhang, Quoc Le, and Mohammad Norouzi. 2021. Speechstew: Simply mix all available speech recognition data to train one large neural network. arXiv preprint, https://arxiv.org/abs/2104.02133. — [arXiv:2104.02133](https://arxiv.org/abs/2104.02133) ↗
  15. William Chen, Wangyou Zhang, Yifan Peng, Xinjian Li, Jinchuan Tian, Jiatong Shi, Xuankai Chang, Soumi Maiti, Karen Livescu, and Shinji Watanabe. 2024. Towards robust speech representation learning for thousands of languages. arXiv pre-print, https://arxiv.org/abs/2407.00837. ↗
  16. Tri Dao, Daniel Y. Fu, Stefano Ermon, Atri Rudra, and Christopher Ré. 2022. FlashAttention: Fast and memory-efficient exact attention with IO-awareness. In Proc. NEURIPS, pages 16344 – 16359. find
  17. Linhao Dong, Shuang Xu, and Bo Xu. 2018. Speech-transformer: A no-recurrence sequence-to-sequence model for speech recognition. In Proc. ICASSP, pages 5884–5888. find
  18. Tomaž Erjavec, Maciej Ogrodniczuk, Petya Osenova, Nikola Ljubešić, Kiril Simov, Andrej Pančur, Michał Rudolf, Matyáš Kopp, Starkaður Barkarson, Steinþór Steingrímsson, Çağrı Çöltekin, Jesse de Does, Katrien Depuydt, Tommaso Agnoloni, Giulia Venturi, María Calzada Pérez, Luciana D. de Macedo, Costanz… find
  19. Sidney Evaldo Leal, Arnaldo Candido Junior, Ricardo Marcacini, Edresson Casanova, Odilon Gonçalves, Anderson Silva Soares, Rodrigo Freitas Lima, Lucas Rafael Stefanel Gris, and Sandra Aluísio. 2025. MuPe life stories dataset: Spontaneous speech in Brazilian Portuguese with a case study evaluation on… find
  20. Marco Gaido, Sara Papi, Luisa Bentivogli, Alessio Brutti, Mauro Cettolo, Roberto Gretter, Marco Matassoni, Mohamed Nabih, and Matteo Negri. 2024. MOSEL: 950,000 hours of speech data for open-source speech foundation model training on EU languages. In Proc. EMNLP, pages 13934–13947, Miami, Florida, U… find
  21. Gonçal Garcés Díaz-Munío, Joan Albert Silvestre Cerdà, Javier Jorge-Cano, Adrián Giménez Pastor, Javier Iranzo-Sánchez, Pau Baquero-Arnal, Nahuel Roselló, Alejandro Manuel Pérez-González de Martos, Jorge Civera Saiz, José Alberto Sanchis Navarro, and Juan Alfons. 2021. Europarl-ASR: A large corpus o… find
  22. Proc. Interspeech 2021, pages 3695–3699. find
  23. Anmol Gulati, James Qin, Chung-Cheng Chiu, Niki Parmar, Yu Zhang, Jiahui Yu, Wei Han, Shibo Wang, Zhengdong Zhang, Yonghui Wu, and Ruoming Pang. 2020. Conformer: Convolution-augmented transformer for speech recognition. In Proc. Interspeech. find our card
  24. Astrid Hagen and Joao P Neto. 2003. HMM/MLP hybrid speech recognizer for the Portuguese telephone SpeechDat corpus. In Proc. PROPOR, pages 126–134. find
  25. J.H.L. Hansen, Rongqing Huang, Bowen Zhou, M. Seadle, J.R. Deller, A.R. Gurijala, M. Kurimo, and P. Angkititrakul. 2005. Speechfind: Advances in spoken document retrieval for a national gallery of the spoken word. IEEE Transactions on Speech and Audio Processing, 13(5):712–730. find
  26. Harald Hoge, Herbert S Tropf, Richard Winski, Henk van den Heuvel, Reinhold Haeb-Umbach, and Khalid Choukri. 1997. European speech databases for telephone applications. In Proc. ICASSP, volume 3, pages 1771–1774. IEEE. find
  27. Javier Iranzo-Sánchez, Joan Albert Silvestre-Cerdà, Javier Jorge, Nahuel Roselló, Adrià Giménez, Albert Sanchis, Jorge Civera, and Alfons Juan. 2020. Europarl-ST: A multilingual corpus for speech translation of parliamentary debates. In Proc. find
  28. Wei Kang, Xiaoyu Yang, Zengwei Yao, Fangjun Kuang, Yifan Yang, Liyong Guo, Long Lin, and Daniel Povey. 2024. Libriheavy: a 50,000 hours asr corpus with punctuation casing and context. In Proc. ICASSP. find
  29. Shigeki Karita, Nanxin Chen, Tomoki Hayashi, Takaaki Hori, Hirofumi Inaguma, Ziyan Jiang, Masao Someki, Nelson Enrique Yalta Soplin, Ryuichi Yamamoto, Xiaofei Wang, Shinji Watanabe, Takenori Yoshimura, and Wangyou Zhang. 2019. A comparative study on transformer vs RNN in speech applications. In Proc… find
  30. Kwangyoun Kim, Felix Wu, Yifan Peng, Jing Pan, Prashant Sridhar, Kyu J. Han, and Shinji Watanabe. 2023. E-Branchformer: Branchformer with enhanced merging for speech recognition. In Proc. SLT. find
  31. Andreas Kirkedal, Marija Stepanović, and Barbara Plank. 2020. FT Speech: Danish parliament speech corpus. In Proc. Interspeech, pages 442– 446. find
  32. Philipp Koehn. 2005. Europarl: A parallel corpus for statistical machine translation. In Proceedings of Machine Translation Summit X: Papers, pages 79–86, Phuket, Thailand. find
  33. Baybars Kulebi, Carme Armentano-Oller, Carlos Rodriguez-Penagos, and Marta Villegas. 2022. ParlamentParla: A speech corpus of Catalan parliamentary sessions. In Proceedings of the Workshop ParlaCLARIN III within the 13th Language Resources and Evaluation Conference, pages 125–130, Marseille, France.… find
  34. Xinjian Li, Shinnosuke Takamichi, Takaaki Saeki, William Chen, Sayaka Shiota, and Shinji Watanabe. 2023. Yodas: Youtube-oriented dataset for audio and speech. In Proc. ASRU, pages 1–8. find
  35. Rodrigo Lima, Sidney E. Leal, Arnaldo Candido Junior, and Sandra M. Aluísio. 2025. A large dataset of spontaneous speech with the accent spoken in São Paulo for automatic speech recognition evaluation. In Proc. Intelligent Systems: 34th Brazilian Conference (BRACIS), pages 33– 47. find
  36. Nikola Ljubešić, Peter Rupnik, and Danijel Koržinek. 2024. The parlaspeech collection of automatically generated speech and text datasets from parliamentary proceedings. In International Conference on Speech and Computer, pages 137– 150. Springer. find
  37. Maria Helena Mateus and Ernesto d’Andrade. 2000. The Phonology Of Portuguese. Oxford University Press. find
  38. Hugo Meinedo, Nuno. Souto, and João P. Neto. 2001. Speech recognition of broadcast news for the European Portuguese language. In Proc. ASRU, pages 319–322. find
  39. Abdelrahman Mohamed, Hung-yi Lee, Lasse Borgholt, Jakob D. Havtorn, Joakim Edin, Christian Igel, Katrin Kirchhoff, Shang-Wen Li, Karen Livescu, Lars Maaløe, Tara N. Sainath, and Shinji Watanabe. 2022. Self-supervised speech representation learning: A review. IEEE Journal of Selected Topics in Signal… find
  40. João Manuel Alves Mourão de Sá. 2021. Reconhecimento de fala em português de Portugal num contexto com poucos recursos. Master’s thesis, Faculdade de Ciências, Universidade do Porto, Porto, Portugal, November. Available at https://hdl.handle.net/10216/139258. ↗
  41. Joao P. Neto, Ciro Martins, and Luis B. Almeida. 1998. A large vocabulary continuous speech recognition hybrid system for the portuguese language. In 5th International Conference on Spoken Language Processing (ICSLP 1998), page paper 0562. find
  42. João P. Neto, Ciro Martins, Hugo Meinedo, and Luis B Almeida. 1997. The design of a large vocabulary speech corpus for Portuguese. In Proc. Eurospeech, pages 1707–1710. find
  43. João P. Neto, Hugo Meinedo, Márcio Viveiros, Renato Cassaca, Ciro Martins, and Diamantino Caseiro. 2008. Broadcast news subtitling system in Portuguese. In Proc. ICASSP, pages 1561– 1564. find
  44. Vassil Panayotov, Guoguo Chen, Daniel Povey, and Sanjeev Khudanpur. 2015. Librispeech: An ASR corpus based on public domain audio books. In Proc. ICASSP. find
  45. Yifan Peng, Shakeel Muhammad, Yui Sudo, William Chen, Jinchuan Tian, Chyi-Jiunn Lin, and Shinji Watanabe. 2025. OWSM v4: Improving open whisper-style speech models via data scaling and cleaning. In Proc Interspeech. find
  46. Yifan Peng, Yui Sudo, Muhammad Shakeel, and Shinji Watanabe. 2024. OWSM-CTC: An Open Encoder-Only Speech Foundation Model for Speech Recognition, Translation, and Language Identification. In Proc. ACL (Volume 1: Long Papers), pages 10192–10209. find
  47. Samuel Pfisterer, Florian Grötschla, Luca Lanzendörfer, Florian Yan, and Roger Wattenhofer. 2025. Eurospeech: A multilingual speech corpus. Proc. NeurIPS. find
  48. Vineel Pratap, Qiantong Xu, Anuroop Sriram, Gabriel Synnaeve, and Ronan Collobert. 2020. Mls: A large-scale multilingual dataset for speech research. In Proc. Interspeech, pages 2757– 2761. find our card
  49. Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, and Ilya Sutskever. 2023. Robust speech recognition via large-scale weak supervision. In Proc. ICML. find our card
  50. Dima Rekesh, Nithin Rao Koluguri, Samuel Kriman, Somshubra Majumdar, Vahid Noroozi, He Huang, Oleksii Hrinchuk, Krishna Puvvada, Ankur Kumar, Jagadeesh Balam, and Boris Ginsburg. 2023. Fast conformer with linearly scalable attention for efficient speech recognition. In Proc. ASRU, pages 1–8. find
  51. Jean-Luc Rouas, Isabel Trancoso, Céu Viana, and Mónica Abreu. 2008. Language and variety verification on broadcast news for portuguese. Speech Communication, 50(11):965–979. find
  52. Temple F. Smith and Michael S. Waterman. 1981. Identification of common molecular subsequences. Journal of Molecular Biology, 147(1):195–197. find
  53. Per Erik Solberg and Pablo Ortiz. 2022. The Norwegian parliamentary speech corpus. In Proc. LREC, pages 1003–1008, Marseille, France. European Language Resources Association. find
  54. Jianlin Su, Murtadha Ahmed, Yu Lu, Shengfeng Pan, Wen Bo, and Yunfeng Liu. 2024. RoFormer: Enhanced transformer with Rotary Position Embedding. Neurocomputing, 568:127063. find
  55. Isabel Trancoso, Joao P. Neto, Hugo Meinedo, and Rui Amaral. 2003. Evaluation of an alert system for selective dissemination of broadcast news. In Proc. Interspeech, pages 1257–1260. find
  56. Shinji Watanabe, Takaaki Hori, Shigeki Karita, Tomoki Hayashi, Jiro Nishitoba, Yuya Unno, Nelson Enrique Yalta Soplin, Jahn Heymann, Matthew Wiesner, Nanxin Chen, Adithya Renduchintala, and Tsubasa Ochiai. 2018. Espnet: End-to-end speech processing toolkit. In Proc. Interspeech. find
  57. Yu Zhang, Wei Han, James Qin, Yongqiang Wang, Ankur Bapna, Zhehuai Chen, Nanxin Chen, Bo Li, Vera Axelrod, Gary Wang, Zhong Meng, Ke Hu, Andrew Rosenberg, Rohit Prabhavalkar, Daniel S. Park, Parisa Haghani, Jason Riesa, Ginger Perng, Hagen Soltau, Trevor Strohman, Bhuvana Ramabhadran, Tara Sainath,… — [arXiv:2303.01037](https://arxiv.org/abs/2303.01037) find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/teixeira-2026-falar-large-scale-speaker-annotated/