Pith. sign in

REVIEW 4 major objections 5 minor 63 references

LIMBA: An Open-Source Framework for the Preservation and Valorization of Low-Resource Languages using Generative Models

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper proposes an open-source, end-to-end framework (LIMBA) for generating data and training a language model in a low-resource language, demonstrated on Sardinian.

desk verdict A clearly written white paper for an integrated Sardinian language-tool pipeline, but the 'open-source framework' claim outruns what is actually built, and the core bootstrap — synthetic data training a generative model — is untested. read the letter →

arxiv 2411.13453 v1 pith:QSA3HK4R submitted 2024-11-20 cs.CL cs.AI

Salvatore Mario Carta , Stefano Chessa , Giulia Contu , Andrea Corriga , Andrea Deidda , Gianni Fenu , Luca Frigau , Alessandro Giuliani
show 14 more authors
Luca Grassi Marco Manolo Manca Mirko Marras Francesco Mola Bastianino Mossa Piergiorgio Mura Marco Ortu Leonardo Piano Simone Pisano Alessia Pisu Alessandro Sebastian Podda Livio Pompianu Simone Seu Sandro Gabriele Tiddia
This is my paper · ORCID
classification cs.CLcs.AI
keywords low-resourcelanguagesSardinianlanguagepreservationgenerativemodelsdatagenerationPart-of-Speechtaggingspeech-to-textmodelfine-tuning
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proposes LIMBA, an open-source framework for building data and training a language model in a low-resource language, using Sardinian as the test case. The central claim is that by chaining data collection, linguistic annotation, machine translation, speech processing, and generative modeling, and filtering everything through a variant identifier and quality checker, a small seed of human-annotated material can grow into a corpus large enough to train a dedicated model. The paper reports preliminary results: a Part-of-Speech tagger at about 89% F1 on roughly 200 test sentences, and a Whisper-based speech-to-text model at about 17 Word Error Rate on roughly two hours of audio. A sympathetic reader would care because the framework is meant to be transferable to other endangered languages, which currently have almost no digital resources.

What carries the argument

The central object is the five-block pipeline architecture: data collection, linguistic modeling (Part-of-Speech tagging), machine translation, speech processing (speech-to-text and text-to-speech), and generative modeling, with two cross-module filters — a Language and Variant Identifier and a Language Quality Checker — that gate which data enters training. The loop is carried by the fine-tuning of pre-trained models on tiny seed data: BERT for token classification and text classification, Whisper for speech-to-text, and open-source LLMs such as Llama3 for translation and generation.

What would settle it

Have native Sardinian speakers or trained linguists rate a random sample of the machine-generated Sardinian sentences for grammaticality and naturalness; if the error rate is high (e.g., above a few percent per sentence), the synthetic loop would contaminate the training corpus rather than expand it. Alternatively, train two Sardinian language models — one on human-only data and one on synthetic-expanded data — and compare perplexity and human ratings; if the synthetic-expanded model is not better, the core claim of the data-generation loop fails.

Watch

Extended reading notes

Core claim

The paper claims that no end-to-end pipeline currently describes how to build new data and train a language model for an under-resourced language, and that LIMBA is a working template for that process. The load-bearing idea is a generative data loop: fine-tune pre-trained models (BERT for tagging and identification, Whisper for transcription, Llama3-style LLMs for translation and generation) on a small human-annotated seed corpus; use those models to produce synthetic Sardinian text and audio; keep only the output that passes the variant identifier and quality checker; then train a dedicated Sardinian language model on the expanded corpus. The paper is explicit that the methodology is still under development, so the contribution is the framework and its early validation numbers rather than a finished Sardinian model.

Load-bearing premise

The load-bearing premise is that a few hundred human-annotated sentences and a couple of hours of audio are enough to fine-tune pre-trained models into tools whose synthetic output stays accurate enough to train a final language model without amplifying errors.

Editorial extensions

If this is right

  • If the methodology works, any language with a few hundred annotated sentences and a few hours of transcribed audio can get a first generation of AI tools, not just Sardinian.
  • The Sardinian PoS tagger, translator, and speech recognizer would be the first automatic tools for the language, giving linguists and standardization efforts concrete analysis resources.
  • The variant identifier and quality checker are designed to keep dialect diversity visible and to prevent low-quality synthetic text from contaminating the corpus.
  • Once enough data is generated, a dedicated Sardinian language model is expected to outperform general LLMs like GPT-4 and Llama3, which the paper says produce grammatically incorrect Sardinian.
  • The open-source release lets speaker communities and researchers adapt the pipeline to other endangered languages rather than waiting for large platforms.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not demonstrate that synthetic data actually improves the final model; a control experiment comparing a model trained on human-only versus synthetic-expanded data would settle that.
  • The bootstrap threshold is unknown: the 700-sentence and 2-hour numbers are what the authors had, not what they established as sufficient, so the minimum viable seed size for other languages is an open question.
  • The variant-aware filtering may be the most fragile part: if the seed corpus under-represents one dialect, the loop could silently erase it while appearing to preserve 'Sardinian,' so dialect coverage should be a reporting metric.
  • Community annotation effort, not model compute, is likely the true bottleneck for transferring LIMBA to another endangered language.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper presents LIMBA, an open-source framework proposal for low-resource language preservation, with Sardinian as a case study. The framework chains five modules—data collection, linguistic modeling (focusing on PoS tagging), machine translation, speech processing (ASR and TTS), and generative modeling—together with two cross-cutting components (language/variant identification and quality checking). The authors report preliminary component-level results: a BERT-based PoS tagger with F1 around 89% on roughly 200 held-out Sardinian sentences, and a fine-tuned Whisper model with a best WER around 17 on about two hours of Common Voice audio. The paper does not report translation scores, does not evaluate the quality of machine-generated Sardinian data, and does not train or evaluate a final language model. The central assertion, stated in Section 1, is that this methodology, still under development, will generate tools capable of constructing new data in little-used languages so that a language model can be trained.

Significance. If the framework worked as claimed, LIMBA would be a useful contribution: a modular, reproducible blueprint for building NLP tools and a generative language model for an endangered language, demonstrated on Sardinian. The paper sensibly identifies dialectal fragmentation, lack of corpora, and absence of a dedicated LM as key bottlenecks, and it connects the technical pipeline to community and standardization concerns. Credit is due for the concrete expert annotation effort (roughly 700 sentences), the use of open models (BERT, Whisper, Llama3), and the explicit acknowledgment that the system is under development. However, the empirical support is preliminary and does not yet substantiate the end-to-end claim. The component scores have no error bars, no baselines, and no data release, and no evidence is provided about the quality, volume, or error-propagation behavior of the synthetic corpus that would be used to train the final language model. The paper therefore reads more as a research program description than as a validated framework, and the experimental gaps are load-bearing for the central claim.

major comments (4)
  1. [§3.6.2 and §1] The central claim that the pipeline produces 'high-quality material' for training a generative model is not supported by any evaluation of the machine-generated corpus. The reported component scores (PoS F1≈89 in §3.3.5, WER≈17 in §3.5.5) are evaluated on human-annotated or human-transcribed data, not on the synthetic text, transcribed audio, or translated sentences that would actually feed the LLM. No volume of generated data, no human-acceptability rate, and no comparison against curated Sardinian text is provided. The assumption that errors from Whisper, Llama3, and the PoS tagger do not amplify when chained is therefore untested. A revision should add a direct evaluation of the generated corpus and an error-propagation analysis.
  2. [§3.4.4 and §3.4.5] The machine translation module, which is described as a primary data-generation engine, reports no experimental results. Section 3.4.5 lists BLEU, TER, and METEOR as metrics, but no scores are given, and §3.4.4 states that experiments are 'currently being tested.' Since translation is the main mechanism for increasing the volume of Sardinian text, the absence of any translation-quality evidence leaves the data-construction claim unsubstantiated for this component.
  3. [§3.3.5 and §3.5.5] The reported evaluation numbers lack the statistical detail needed to support the claim of 'excellent values.' PoS F1 of about 89% is a single point estimate on roughly 200 test sentences, with no confidence interval, no variance across runs, and no comparison against a baseline (e.g., zero-shot multilingual BERT, a majority-class model, or earlier Sardinian POS efforts). Whisper's WER of about 17 is reported as the 'smallest value' without a confidence interval or a clear description of the test set. The revision should provide error bars, baselines, and explicit descriptions of the test sets and splits.
  4. [§3.3.3 and title] The paper calls the framework 'open-source' but does not release the ~700-sentence PoS dataset, the audio data, the fine-tuned models, or the code. For a contribution whose value lies in a reproducible methodology for endangered languages, the lack of a public repository or data-link makes the title's 'open-source' claim unverifiable. Please provide a link or clearly state the data-availability plan.
minor comments (5)
  1. [Title and abstract] The title reads 'A OPEN-SOURCE' and should be 'An Open-Source'; Section 5 heading contains the typo 'Conlusions' instead of 'Conclusions.'
  2. [§3.5.1, Eq. (5)] The sentence describing Eq. (5) says 'commonly known as speech-to-text' but the equation is for text-to-speech; the label should be corrected.
  3. [Table 2 (Section 3.5.4)] The table lists 'Training Loss Word Error Rate,' but WER is an evaluation metric, not a differentiable loss function. The actual training loss (e.g., CTC or cross-entropy) should be stated.
  4. [§3.2.1 and §3.6.4] The language/variant identifier and the final LLM are described as 'underway' or 'currently being analyzed,' but no commitment is made about when results will be added. Since these components are essential to the framework, a clearer statement of the current status and the planned evaluation would help the reader situate the contribution.
  5. [References] Some references have inconsistent formatting, e.g., [36] lists an incomplete author string and [21] lists the Apertium paper with an unusual author order; please standardize according to the bibliography style.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the framework is a proposed methodology with component-level evaluations; self-citations are motivational, not load-bearing.

full rationale

This is a methodology white paper rather than a derivation, and no load-bearing step reduces to its own inputs. The formal sections (Eqs. 1-8) define classification, translation, speech, and generative objectives; they do not fit parameters and then relabel those fits as predictions. The only reported numbers (PoS F1 approximately 89% on roughly 200 held-out sentences; Whisper WER approximately 17 on about two hours of Common Voice audio) are component-level evaluations, and the paper neither uses them as proof of end-to-end success nor claims that a final Sardinian language model has already been produced. Self-citations [39] and [46] describe the authors' ongoing PoS tagger and Sardinian corpus; they are cited to document the current absence of tools and the existence of seed resources, so the framework's design does not depend on those results being true. The 'no end-to-end pipelines' statement is a literature assertion, not a uniqueness theorem imported from the authors' prior work. The main weakness is empirical: the quality and error propagation of machine-generated Sardinian data are never measured, and several modules are explicitly said to be under development or still being analyzed. That is an unvalidated assumption and a completeness risk, but not circularity.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free parameters or invented entities are introduced. The framework rests on domain-level assumptions about data scarcity, transfer learning, and evaluation metrics, which are plausible but not demonstrated.

assumptions (4)
  • domain assumption Sardinian is an endangered language with insufficient digital resources to train language models.
    Motivates the whole work in the abstract and Section 1; supported by cited literature but is a factual premise about the language.
  • ad hoc to paper Fine-tuning small pre-trained multilingual models on limited expert-annotated and generated Sardinian data yields reliable language tools.
    Central working hypothesis in Sections 3.3.4, 3.4.4, 3.5.4 and 3.6.4; only preliminary evidence is given.
  • domain assumption Automatic metrics (F1, WER, BLEU, perplexity) adequately measure the quality of tools for language preservation.
    Invoked in the evaluation sections (3.3.5, 3.5.5, 3.6.5) without a discussion of their limits for endangered-language preservation.
  • domain assumption The collected Common Voice Sardinian subset and expert annotations are representative of the language's macro-variants.
    Common Voice data is described in Section 3.5.2 as providing about two hours of Sardinian in multiple macro-variants; no variant distribution or quality audit is given.

how reviews work

0 comments
Cite this review

Pith. "Pith review of LIMBA: An Open-Source Framework for the Preservation and Valorization of Low-Resource Languages using Generative Models." pith.science (2026). https://pith.science/paper/QSA3HK4R

@misc{pith2026241113453,
  author       = {Pith},
  title        = {Pith review of: LIMBA: An Open-Source Framework for the Preservation and Valorization of Low-Resource Languages using Generative Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QSA3HK4R}},
  note         = {Machine review of arXiv:2411.13453}
}
read the original abstract

Minority languages are vital to preserving cultural heritage, yet they face growing risks of extinction due to limited digital resources and the dominance of artificial intelligence models trained on high-resource languages. This white paper proposes a framework to generate linguistic tools for low-resource languages, focusing on data creation to support the development of language models that can aid in preservation efforts. Sardinian, an endangered language, serves as the case study to demonstrate the framework's effectiveness. By addressing the data scarcity that hinders intelligent applications for such languages, we contribute to promoting linguistic diversity and support ongoing efforts in language standardization and revitalization through modern technologies.

Figures

Figures reproduced from arXiv: 2411.13453 by the authors.

Figure 1
Figure 1. depicts the architecture of the developed framework, which is characterized by the concatenation of multiple modules, each producing a result in its own right. Hence, in addition to providing a method for generating a language model for low-resource languages, AI-based tools suitable for developing and enhancing the language of interest are described and developed [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

63 extracted references · 51 canonical work pages

  1. [1]

    Common voice: A massively-multilingual speech corpus

    Rosana Ardila, Megan Branson, Kelly Davis, Michael Henretty, Michael Kohler, Josh Meyer, Reuben Morais, Lindsay Saunders, Francis M Tyers, and Gregor Weber. Common voice: A massively-multilingual speech corpus. arXiv preprint arXiv:1912.06670, 2019

    arXiv 1912

  2. [2]

    Introducing Syllable Tokenization for Low-resource Languages: A Case Study with Swahili

    Jesse Atuhurra, Hiroyuki Shindo, Hidetaka Kamigaito, and Taro Watanabe. Introducing syllable tokenization for low-resource languages: A case study with swahili. arXiv preprint arXiv:2406.15358, 2024

    work page Pith review arXiv 2024

  3. [3]

    Low-Resource Speech-to-Text Translation

    Sameer Bansal, Herman Kamper, Karen Livescu, Adam Lopez, and Sharon Goldwater. Low-resource speech-to- text translation. arXiv preprint arXiv:1803.09164, 2018

    work page Pith review arXiv 2018

  4. [4]

    Language Model Prior for Low-Resource Neural Machine Translation

    Christos Baziotis, Barry Haddow, and Alexandra Birch. Language model prior for low-resource neural machine translation. arXiv preprint arXiv:2004.14928, 2020

    work page Pith review arXiv 2004

  5. [5]

    Decolonising speech and language technology

    Steven Bird. Decolonising speech and language technology. In 28th International Conference on Computational Linguistics, COLING 2020, pages 3504–3519. Association for Computational Linguistics (ACL), 2020

    work page 2020

  6. [6]

    Cross-Lingual Morphological Tagging for Low-Resource Languages

    Jan Buys and Jan A Botha. Cross-lingual morphological tagging for low-resource languages. arXiv preprint arXiv:1606.04279, 2016

    work page Pith review arXiv 2016

  7. [7]

    Llms are few-shot in-context low-resource language learners

    Samuel Cahyawijaya, Holy Lovenia, and Pascale Fung. Llms are few-shot in-context low-resource language learners. arXiv preprint arXiv:2403.16512, 2024

    arXiv 2024

  8. [8]

    A Grounded Unsupervised Universal Part-of-Speech Tagger for Low-Resource Languages

    Ronald Cardenas, Ying Lin, Heng Ji, and Jonathan May. A grounded unsupervised universal part-of-speech tagger for low-resource languages. arXiv preprint arXiv:1904.05426, 2019

    work page Pith review arXiv 1904

Show all 63 references
  1. [9]

    Two decades of unsupervised pos induction: How far have we come? In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 575–584, 2010

    Christos Christodoulopoulos, Sharon Goldwater, and Mark Steedman. Two decades of unsupervised pos induction: How far have we come? In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 575–584, 2010

    2010

  2. [10]

    Selection criteria for low resource language programs

    Christopher Cieri, Mike Maxwell, Stephanie Strassel, and Jennifer Tracey. Selection criteria for low resource language programs. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16), pages 4543–4549, 2016

    2016

  3. [11]

    Data collection pipeline for low-resource languages: A case study on constructing a tetun text corpus

    Gabriel de Jesus and Sérgio Sobral Nunes. Data collection pipeline for low-resource languages: A case study on constructing a tetun text corpus. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLI...

    2024

  4. [12]

    Bert: Pre-training of deep bidirectional transformers for language understanding

    Jacob Devlin. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805, 2018

    2018 arXiv

  5. [13]

    Statistical identification of language

    Ted Dunning. Statistical identification of language. Computing Research Laboratory, New Mexico State University Las Cruces, 1994

    1994

  6. [14]

    What can we get from 1000 tokens? a case study of multilingual pos tagging for resource-poor languages

    Long Duong, Trevor Cohn, Karin Verspoor, Steven Bird, and Paul Cook. What can we get from 1000 tokens? a case study of multilingual pos tagging for resource-poor languages. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages...

    2014

  7. [15]

    Learning when to trust distant supervision: An application to low-resource pos tagging using cross-lingual projection

    Meng Fang and Trevor Cohn. Learning when to trust distant supervision: An application to low-resource pos tagging using cross-lingual projection. arXiv preprint arXiv:1607.01133, 2016

    2016 arXiv

  8. [16]

    Model transfer for tagging low-resource languages using a bilingual dictionary

    Meng Fang and Trevor Cohn. Model transfer for tagging low-resource languages using a bilingual dictionary. arXiv preprint arXiv:1705.00424, 2017

    2017 arXiv

  9. [17]

    Manuale di linguistica sarda

    Eduardo Blasco Ferrer, Peter Koch, and Daniela Marzo, editors. Manuale di linguistica sarda. De Gruyter, Berlin, Boston, 2017. ISBN 9783110274615. doi:doi:10.1515/9783110274615. URL https://doi.org/10.1515/ 9783110274615

    2017 doi

  10. [18]

    Ideologies and expressed attitudes in internet: Comparing ethnic identities in two regional communities (veneto and sardinia)

    Anna Ghimenton and Giovanni Depau. Ideologies and expressed attitudes in internet: Comparing ethnic identities in two regional communities (veneto and sardinia). Vanishing Languages in Context. Ideological, Attitudinal and Social Identity Perspectives, 114:73–102, 2016

    2016

  11. [19]

    Metrics for multi-class classification: an overview

    Margherita Grandini, Enrico Bagli, and Giorgio Visani. Metrics for multi-class classification: an overview. arXiv preprint arXiv:2008.05756, 2020

    2008 arXiv

  12. [20]

    Speech generation for indigenous language education

    Ross Krekoskid Kazantsevaa, Roland Kuhna, Samuel Larkina, Patrick Littella, Delaney Lothiana, Korin Rich- mondc Akwiratékha’Martina, Marc Tessiera, Cassia Valentini-Botinhaoc, Dan Wellsc, and Junichi Yamagishib. Speech generation for indigenous language education. Computer Spe...

    2024

  13. [21]

    Recent advances in apertium, a free/open-source rule-based machine translation platform for low-resource languages

    Tanmai Khanna, Jonathan N Washington, Francis M Tyers, Sevilay Bayatlı, Daniel G Swanson, Tommi A Pirinen, Irene Tang, and Hector Alos i Font. Recent advances in apertium, a free/open-source rule-based machine translation platform for low-resource languages. Machine Translatio...

    2021

  14. [22]

    Practical Natural Language Processing for Low-Resource Languages

    Benjamin Philip King. Practical Natural Language Processing for Low-Resource Languages. PhD thesis, University of Michigan, 2015

    2015

  15. [23]

    Synthesizer voice quality of new languages calibrated with mean mel cepstral distortion

    John Kominek, Tanja Schultz, and Alan W Black. Synthesizer voice quality of new languages calibrated with mean mel cepstral distortion. In SLTU, pages 63–68, 2008

    2008

  16. [24]

    An augmented translation technique for low resource language pair: Sanskrit to hindi translation

    Rashi Kumar, Piyush Jha, and Vineet Sahula. An augmented translation technique for low resource language pair: Sanskrit to hindi translation. In Proceedings of the 2019 2nd international conference on algorithms, computing and artificial intelligence, pages 377–383, 2019

    2019

  17. [25]

    adaptmllm: Fine-tuning multilingual language models on low-resource languages with integrated llm playgrounds

    Séamus Lankford, Haithem Afli, and Andy Way. adaptmllm: Fine-tuning multilingual language models on low-resource languages with integrated llm playgrounds. Information, 14(12):638, 2023

    2023

  18. [26]

    A survey on evaluation metrics for machine translation

    Seungjun Lee, Jungseob Lee, Hyeonseok Moon, Chanjun Park, Jaehyung Seo, Sugyeong Eo, Seonmin Koo, and Heuiseok Lim. A survey on evaluation metrics for machine translation. Mathematics, 11(4):1006, 2023

    2023

  19. [27]

    Laura Linzmeier, Simone Pisano, et al. Visibilità delle varietà italo-romanze nel paesaggio linguistico della sardegna settentrionale e nel cyberspazio: il caso del sassarese e del gallurese.La presenza dei dialetti italo-romanzi nel paesaggio linguistico. Ricerche e Riflessio...

    2021

  20. [28]

    Practical comparable data collection for low-resource languages via images

    Aman Madaan, Shruti Rijhwani, Antonios Anastasopoulos, Yiming Yang, and Graham Neubig. Practical comparable data collection for low-resource languages via images. arXiv preprint arXiv:2004.11954, 2020

    2004 arXiv

  21. [29]

    Low-resource languages: A review of past work and future challenges

    Alexandre Magueresse, Vincent Carles, and Evan Heetderks. Low-resource languages: A review of past work and future challenges. arXiv preprint arXiv:2006.07264, 2020

    2006 arXiv

  22. [30]

    Neural transition-based string transduction for limited-resource setting in morphology

    Peter Makarov and Simon Clematide. Neural transition-based string transduction for limited-resource setting in morphology. In Proceedings of the 27th International Conference on Computational Linguistics, pages 83–93, 2018

    2018

  23. [31]

    Lingue di minoranza a scuola: uno sguardoalla sardegna a dieci anni dalla legge 482/99

    Maria Antonietta Marra. Lingue di minoranza a scuola: uno sguardoalla sardegna a dieci anni dalla legge 482/99. In Linguistica educativa: atti del XLIV Congresso internazionale di studi della Società di linguistica italiana (SLI): Viterbo, 27-29 settembre 2010.-(Pubblicazioni ...

    2010

  24. [32]

    Dialetti su instagram: usi, differenze e atteggiamenti linguistici

    Camilla Masullo, Claudia Castelli, Cinzia Meloni, Chiara Meluzzi, et al. Dialetti su instagram: usi, differenze e atteggiamenti linguistici. BIBLIOTECA DI LINGUISTICA E FILOLOGIA, 7:237–254, 2021

    2021

  25. [33]

    The internet as a rescue tool of endangered languages: Sardinian

    Guido Mensching. The internet as a rescue tool of endangered languages: Sardinian. In Proceeding Conference Multilinguae: multimedia and minority languages. San Sebastian: The Association of Electronics and Information Technology Industries, 2000

    2000

  26. [34]

    Class-based n-gram models of natural language

    Robert L Mercer et al. Class-based n-gram models of natural language. Computational Linguistics, 18(4):18–4, 1992

    1992

  27. [35]

    Efforts to standardise minority languages

    Daniela Mereu. Efforts to standardise minority languages. the case of sardinian. Europäisches Journal für Minderheitenfragen, 14(1-2):76–95, 2021

    2021

  28. [36]

    Merlo and Scuola normale superiore (Italy)

    C. Merlo and Scuola normale superiore (Italy). L’Italia dialettale: rivista di dialettologia italiana . L’Italia dialettale. Arti Grafiche Pacini Mariotti, 2009. URL https://books.google.it/books?id=3h_DAr8XOOEC

    2009

  29. [37]

    Applying AI-based Tools and Technologies Towards Revitalization of Indigenous and Endangered Languages

    Sushree Sangita Mohanty, Satya Ranjan Dash, and Shantipriya Parida. Applying AI-based Tools and Technologies Towards Revitalization of Indigenous and Endangered Languages. Springer, 2024

    2024

  30. [38]

    Attitudes towards sardinian and italian finally compared via the matched-guise technique

    Piergiorgio Mura. Attitudes towards sardinian and italian finally compared via the matched-guise technique. International Journal of the Sociology of Language, 2024(288):121–147, 2024

    2024

  31. [39]

    The corpus of Sardinian emigrants:a tool for a quantitative approach to contact phenomena

    Piergiorgio Mura, Simone Pisano, Salvatore Carta, Alessandro Giuliani, and Manolo Manca. The corpus of Sardinian emigrants:a tool for a quantitative approach to contact phenomena. MiLES: Minority Languages in European Societies - International Conference-Turin / Bard - BOOK OF...

    2024

  32. [40]

    The valorisation of african languages and policies in the african education systems: A case of uganda

    Michael Muzoora, Daniel R Terry, and Agatha A Asiimwe. The valorisation of african languages and policies in the african education systems: A case of uganda. Universal Journal of Educational Research, 2(1):42–50, 2014

    2014

  33. [41]

    Revitalizing endangered languages: A practical guide

    Justyna Olko and Julia Sallabank. Revitalizing endangered languages: A practical guide. Cambridge University Press, 2021

    2021

  34. [42]

    Building pre-train llm dataset for the indic languages: a case study on hindi

    Shantipriya Parida, Shakshi Panwar, Kusum Lata, Sanskruti Mishra, and Sambit Sekhar. Building pre-train llm dataset for the indic languages: a case study on hindi. arXiv preprint arXiv:2407.09855, 2024

    2024 arXiv

  35. [43]

    An empirical analysis of word error rate and keyword error rate

    Youngja Park, Siddharth Patwardhan, Karthik Visweswariah, and Stephen C Gates. An empirical analysis of word error rate and keyword error rate. In Interspeech, volume 2008, pages 2070–2073, 2008. 17 LIMBA: An Open-Source Framework for the Preservation and Valorization of Low-R...

    2008

  36. [44]

    Towards a more inclusive ai: Progress and perspectives in large language model training for the s \’ami language

    Ronny Paul, Himanshu Buckchash, Shantipriya Parida, and Dilip K Prasad. Towards a more inclusive ai: Progress and perspectives in large language model training for the s \’ami language. arXiv preprint arXiv:2405.05777, 2024

    2024 arXiv

  37. [45]

    Carta dei dialetti d’italia

    Giovan Battista Pellegrini. Carta dei dialetti d’italia. (No Title), 1977

    1977

  38. [46]

    Appunti per un corpus di sardo multimediale

    Simone Pisano, Valentina Piunno, Vittorio Ganfi, et al. Appunti per un corpus di sardo multimediale. In Per una pianificazione del plurilinguismo in Sardegna, pages 147–164. Condaghes, 2022

    2022

  39. [47]

    Applying monte carlo techniques to language identification

    Arjen Poutsma. Applying monte carlo techniques to language identification. In Computational Linguistics in the Netherlands 2001, pages 179–189. Brill, 2002

    2001

  40. [48]

    Robust speech recognition via large-scale weak supervision, 2022

    Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, and Ilya Sutskever. Robust speech recognition via large-scale weak supervision, 2022. URL https://arxiv.org/abs/2212.04356

    2022 arXiv

  41. [49]

    Sardinian on facebook: Analysing diatopic varieties through translated lexical lists

    Irene Russo, Simone Pisano, and Claudia Soria. Sardinian on facebook: Analysing diatopic varieties through translated lexical lists. In CLiC-it/EVALITA, 2016

    2016

  42. [50]

    Analyzing asr pretraining for low-resource speech-to- text translation

    Mihaela C Stoian, Sameer Bansal, and Sharon Goldwater. Analyzing asr pretraining for low-resource speech-to- text translation. In ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 7909–7913. IEEE, 2020

    2020

  43. [51]

    Low-resource taxonomy enrichment with pretrained language models

    Kunihiro Takeoka, Kosuke Akimoto, and Masafumi Oyamada. Low-resource taxonomy enrichment with pretrained language models. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 2747–2758, 2021

    2021

  44. [52]

    End-to-end text-to-speech for low-resource languages by cross-lingual transfer learning

    Tao Tu, Yuan-Jui Chen, Cheng-chieh Yeh, and Hung-Yi Lee. End-to-end text-to-speech for low-resource languages by cross-lingual transfer learning. arXiv preprint arXiv:1904.06508, 2019

    1904 arXiv

  45. [53]

    Rule-based machine translation for the italian-sardinian language pair

    Francis M Tyers, Hèctor Alòs i Font, Gianfranco Fronteddu, and Adrià Martín-Mor. Rule-based machine translation for the italian-sardinian language pair. The Prague Bulletin of Mathematical Linguistics, 108(1):221, 2017

    2017

  46. [54]

    Capitolo secondo: Valori, opinioni e atteggiamenti verso le lingue locali

    M Valdes. Capitolo secondo: Valori, opinioni e atteggiamenti verso le lingue locali. Le lingue dei sardi-Una ricerca sociolinguistica, pages 46–64, 2007

    2007

  47. [55]

    La lingua e la cultura della Sardegna

    Maurizio Virdis. La lingua sarda fra le lingue neolatine. storia uso e problemi. In Convegno “La lingua e la cultura della Sardegna”, 2003. URL https://www.academia.edu/22772692/La_lingua_sarda_fra_le_ lingue_neolatine_Storia_uso_e_problemi

    2003

  48. [56]

    Measuring speech quality for text-to-speech systems: development and assessment of a modified mean opinion score (mos) scale

    Mahesh Viswanathan and Madhubalan Viswanathan. Measuring speech quality for text-to-speech systems: development and assessment of a modified mean opinion score (mos) scale. Computer speech & language, 19(1): 55–83, 2005

    2005

  49. [57]

    Wagner and Rudolf Virchowstiftung

    M.L. Wagner and Rudolf Virchowstiftung. Das ländliche Leben Sardiniens im Spiegel der Sprache: kulturhistorisch-sprachliche Untersuchungen. Wörter und Sachen : Kulturhistorische Zeitschrift für Sprach- und Sachforschung. Beiheft 4. na, 1921. URL https://books.google.it/books?i...

    1921

  50. [58]

    Data selection for language models via importance resampling

    Sang Michael Xie, Shibani Santurkar, Tengyu Ma, and Percy S Liang. Data selection for language models via importance resampling. Advances in Neural Information Processing Systems, 36:34201–34227, 2023

    2023

  51. [59]

    Inducing multilingual text analysis tools via robust projection across aligned corpora

    David Yarowsky, Grace Ngai, and Richard Wicentowski. Inducing multilingual text analysis tools via robust projection across aligned corpora. In Proceedings of the First International Conference on Human Language Technology Research, 2001. URL https://aclanthology.org/H01-1035

    2001

  52. [60]

    Data collection design for dialogue systems for low-resource languages

    Zulipiye Yusupujiang and Jonathan Ginzburg. Data collection design for dialogue systems for low-resource languages. Conversational Dialogue Systems for the Next Decade, pages 387–392, 2021

    2021

  53. [61]

    Machine translation with large language models: Prompting, few-shot learning, and fine-tuning with qlora

    Xuan Zhang, Navid Rajabi, Kevin Duh, and Philipp Koehn. Machine translation with large language models: Prompting, few-shot learning, and fine-tuning with qlora. In Proceedings of the Eighth Conference on Machine Translation, pages 468–481, 2023

    2023

  54. [62]

    Fine-tuning large language models for domain-specific machine translation

    Jiawei Zheng, Hanghai Hong, Xiaoli Wang, Jingsong Su, Yonggui Liang, and Shikai Wu. Fine-tuning large language models for domain-specific machine translation. arXiv preprint arXiv:2402.15061, 2024

    2024 arXiv

  55. [63]

    Transfer learning for low-resource neural machine translation

    Barret Zoph, Deniz Yuret, Jonathan May, and Kevin Knight. Transfer learning for low-resource neural machine translation. arXiv preprint arXiv:1604.02201, 2016. 18

    2016 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.