Towards Best Practices for Open Datasets for LLM Training.

Maarten Van Segbroeck

Stefan Baack Stella Biderman Kasia Odrozek Aviya Skowron Ayah Bdeir Jillian Bommarito Jennifer Ding Maximilian Gahntz Paul Keller Pierre-Carl Langlais Greg Lindahl Sebastian Majstorovic Nik Marda Guilherme Penedo Maarten Van Segbroeck Jennifer Wang Leandro von Werra Mitchell Baker Julie Belião Kasia Chmielinski Marzieh Fadaee Lisa Gutermuth Hynek Kydlícek Greg Leppert EM Lewis-Jong Solana Larsen Shayne Longpre Angela Oduor Lungati Cullen Miller Victor Miller Max Ryabinin Kathleen Siminyu Andrew Strait Mark Surman Anna Tumadóttir Maurice Weber Rebecca Weiss Lee White Thomas Wolf 0008 Towards Best Practices for Open Datasets for LLM Training. 2025 January abs/2501.08365 CoRR https://doi.org/10.48550/arXiv.2501.08365 db/journals/corr/corr2501.html#abs-2501-08365 streams/journals/corr

Liwen You Erika Pelaez Coyotl Suren Gunturu Maarten Van Segbroeck Transformer-Based Bioacoustic Sound Event Detection on Few-Shot Learning Tasks. 1-5 2023 ICASSP https://doi.org/10.1109/ICASSP49357.2023.10097081 conf/icassp/2023 db/conf/icassp/icassp2023.html#YouCGS23 Andreas Schwarz Di He 0004 Maarten Van Segbroeck Mohammed Hethnawi Ariya Rastrow Personalized Predictive ASR for Latency Reduction in Voice Assistants. 745-749 2023 INTERSPEECH https://doi.org/10.21437/Interspeech.2023-211 conf/interspeech/2023 db/conf/interspeech/interspeech2023.html#Schwarz0SHR23

Andreas Schwarz Di He 0004 Maarten Van Segbroeck Mohammed Hethnawi Ariya Rastrow Personalized Predictive ASR for Latency Reduction in Voice Assistants. 2023 abs/2305.13794 CoRR https://doi.org/10.48550/arXiv.2305.13794 db/journals/corr/corr2305.html#abs-2305-13794

Maarten Van Segbroeck Ahmed Zaid Ksenia Kutsenko Cirenia Huerta Tinh Nguyen Xuewen Luo Björn Hoffmeister Jan Trmal Maurizio Omologo Roland Maas DiPCo - Dinner Party Corpus. 434-436 2020 INTERSPEECH https://doi.org/10.21437/Interspeech.2020-2800 conf/interspeech/2020 db/conf/interspeech/interspeech2020.html#SegbroeckZKHNLH20 Jinxi Guo Gautam Tiwari Jasha Droppo Maarten Van Segbroeck Che-Wei Huang Andreas Stolcke Roland Maas Efficient Minimum Word Error Rate Training of RNN-Transducer for End-to-End Speech Recognition. 2807-2811 2020 INTERSPEECH https://doi.org/10.21437/Interspeech.2020-1557 conf/interspeech/2020 db/conf/interspeech/interspeech2020.html#GuoTDSHSM20

Maarten Van Segbroeck Sri Harish Mallidi Brian John King I-Fan Chen Gurpreet Chadha Roland Maas Multi-view Frequency LSTM: An Efficient Frontend for Automatic Speech Recognition. 2020 abs/2007.00131 CoRR https://arxiv.org/abs/2007.00131 db/journals/corr/corr2007.html#abs-2007-00131

Jinxi Guo Gautam Tiwari Jasha Droppo Maarten Van Segbroeck Che-Wei Huang Andreas Stolcke Roland Maas Efficient minimum word error rate training of RNN-Transducer for end-to-end speech recognition. 2020 abs/2007.13802 CoRR https://arxiv.org/abs/2007.13802 db/journals/corr/corr2007.html#abs-2007-13802

Maarten Van Segbroeck Ahmed Zaid Ksenia Kutsenko Cirenia Huerta Tinh Nguyen Xuewen Luo Björn Hoffmeister Jan Trmal Maurizio Omologo Roland Maas DiPCo - Dinner Party Corpus. 2019 abs/1909.13447 CoRR http://arxiv.org/abs/1909.13447 db/journals/corr/corr1909.html#abs-1909-13447

Vikram Ramanarayanan Maarten Van Segbroeck Shrikanth S. Narayanan Directly data-derived articulatory gesture-like representations retain discriminatory information about phone categories. 330-346 2016 36 Comput. Speech Lang. https://doi.org/10.1016/j.csl.2015.03.004 https://www.wikidata.org/entity/Q30362120 db/journals/csl/csl36.html#RamanarayananSN16

Maarten Van Segbroeck Ruchir Travadi Shrikanth S. Narayanan Rapid Language Identification. 1118-1129 2015 23 IEEE ACM Trans. Audio Speech Lang. Process. 7 https://doi.org/10.1109/TASLP.2015.2419978 db/journals/taslp/taslp23.html#SegbroeckTN15

Samuel Thomas 0001 George Saon Maarten Van Segbroeck Shrikanth S. Narayanan Improvements to the IBM speech activity detection system for the DARPA RATS program. 4500-4504 2015 ICASSP https://doi.org/10.1109/ICASSP.2015.7178822 conf/icassp/2015 db/conf/icassp/icassp2015.html#ThomasSSN15 Matthew P. Black Daniel Bone Zisis Iason Skordilis Rahul Gupta 0001 Wei Xia Pavlos Papadopoulos Sandeep Nallan Chakravarthula Bo Xiao 0003 Maarten Van Segbroeck Jangwon Kim Panayiotis G. Georgiou Shrikanth S. Narayanan Automated evaluation of non-native English pronunciation quality: combining knowledge- and data-driven features at multiple time scales. 493-497 2015 INTERSPEECH https://doi.org/10.21437/Interspeech.2015-182 conf/interspeech/2015 db/conf/interspeech/interspeech2015.html#BlackBSGXPCXSKG15 Jangwon Kim Md. Nasir Rahul Gupta 0001 Maarten Van Segbroeck Daniel Bone Matthew P. Black Zisis Iason Skordilis Zhaojun Yang Panayiotis G. Georgiou Shrikanth S. Narayanan Automatic estimation of parkinson's disease severity from diverse speech tasks. 914-918 2015 INTERSPEECH https://doi.org/10.21437/Interspeech.2015-194 conf/interspeech/2015 db/conf/interspeech/interspeech2015.html#KimNGSBBSYGN15 Naveen Kumar 0004 Maarten Van Segbroeck Kartik Audhkhasi Peter Drotár Shrikanth S. Narayanan Fusion of diverse denoising systems for robust automatic speech recognition. 5557-5561 2014 ICASSP https://doi.org/10.1109/ICASSP.2014.6854666 conf/icassp/2014 db/conf/icassp/icassp2014.html#0004SADN14 Bo Xiao 0003 Daniel Bone Maarten Van Segbroeck Zac E. Imel David C. Atkins Panayiotis G. Georgiou Shrikanth S. Narayanan Modeling therapist empathy through prosody in drug addiction counseling. 213-217 2014 INTERSPEECH https://doi.org/10.21437/Interspeech.2014-55 conf/interspeech/2014 db/conf/interspeech/interspeech2014.html#XiaoBSIAGN14 James Gibson Maarten Van Segbroeck Shrikanth S. Narayanan Comparing time-frequency representations for directional derivative features. 612-615 2014 INTERSPEECH https://doi.org/10.21437/Interspeech.2014-147 conf/interspeech/2014 db/conf/interspeech/interspeech2014.html#GibsonSN14 Maarten Van Segbroeck Ruchir Travadi Colin Vaz Jangwon Kim Matthew P. Black Alexandros Potamianos Shrikanth S. Narayanan Classification of cognitive load from speech using an i-vector framework. 751-755 2014 INTERSPEECH https://doi.org/10.21437/Interspeech.2014-114 conf/interspeech/2014 db/conf/interspeech/interspeech2014.html#SegbroeckTVKBPN14 Sriram Ganapathy Kyu Jeong Han Samuel Thomas 0001 Mohamed Kamal Omar Maarten Van Segbroeck Shrikanth S. Narayanan Robust language identification using convolutional neural network features. 1846-1850 2014 INTERSPEECH https://doi.org/10.21437/Interspeech.2014-419 conf/interspeech/2014 db/conf/interspeech/interspeech2014.html#GanapathyHTOSN14 Maarten Van Segbroeck Ruchir Travadi Shrikanth S. Narayanan UBM fused total variability modeling for language identification. 3027-3031 2014 INTERSPEECH https://doi.org/10.21437/Interspeech.2014-607 conf/interspeech/2014 db/conf/interspeech/interspeech2014.html#SegbroeckTN14 Ruchir Travadi Maarten Van Segbroeck Shrikanth S. Narayanan Modified-prior i-vector estimation for language identification of short duration utterances. 3037-3041 2014 INTERSPEECH https://doi.org/10.21437/Interspeech.2014-609 conf/interspeech/2014 db/conf/interspeech/interspeech2014.html#TravadiSN14 Naveen Kumar 0004 Rahul Gupta 0001 Tanaya Guha Colin Vaz Maarten Van Segbroeck Jangwon Kim Shrikanth S. Narayanan Affective Feature Design and Predicting Continuous Affective Dimensions from Music. 2014 MediaEval conf/mediaeval/2014 db/conf/mediaeval/mediaeval2014.html#0004GGVSKN14 https://ceur-ws.org/Vol-1263/mediaeval2014_submission_24.pdf Rahul Gupta 0001 Nikolaos Malandrakis Bo Xiao 0003 Tanaya Guha Maarten Van Segbroeck Matthew Black Alexandros Potamianos Shrikanth S. Narayanan Multimodal Prediction of Affective Dimensions and Depression in Human-Computer Interactions. 33-40 2014 AVEC@MM https://doi.org/10.1145/2661806.2661810 conf/mm/2014avec db/conf/mm/avec2014.html#GuptaMXGSBPN14 Maarten Van Segbroeck Shrikanth S. Narayanan A robust frontend for ASR: Combining denoising, noise masking and feature normalization. 7097-7101 2013 ICASSP https://doi.org/10.1109/ICASSP.2013.6639039 conf/icassp/2013 db/conf/icassp/icassp2013.html#SegbroeckN13 Ming Li 0026 Andreas Tsiartas Maarten Van Segbroeck Shrikanth S. Narayanan Speaker verification using simplified and supervised i-vector modeling. 7199-7203 2013 ICASSP https://doi.org/10.1109/ICASSP.2013.6639060 conf/icassp/2013 db/conf/icassp/icassp2013.html#LiTSN13 Daniel Bone Theodora Chaspari Kartik Audhkhasi James Gibson Andreas Tsiartas Maarten Van Segbroeck Ming Li 0026 Sungbok Lee Shrikanth S. Narayanan Classifying language-related developmental disorders from speech cues: the promise and the potential confounds. 182-186 2013 INTERSPEECH https://doi.org/10.21437/Interspeech.2013-63 conf/interspeech/2013 db/conf/interspeech/interspeech2013.html#BoneCAGTSLLN13 Maarten Van Segbroeck Andreas Tsiartas Shrikanth S. Narayanan A robust frontend for VAD: exploiting contextual, discriminative and spectral cues of human voice. 704-708 2013 INTERSPEECH https://doi.org/10.21437/Interspeech.2013-198 conf/interspeech/2013 db/conf/interspeech/interspeech2013.html#SegbroeckTN13 Andreas Tsiartas Theodora Chaspari Nassos Katsamanis Prasanta Kumar Ghosh Ming Li 0026 Maarten Van Segbroeck Alexandros Potamianos Shrikanth S. Narayanan Multi-band long-term signal variability features for robust voice activity detection. 718-722 2013 INTERSPEECH https://doi.org/10.21437/Interspeech.2013-201 conf/interspeech/2013 db/conf/interspeech/interspeech2013.html#TsiartasCKGLSPN13 James Gibson Maarten Van Segbroeck Antonio Ortega Panayiotis G. Georgiou Shrikanth S. Narayanan Spectro-temporal directional derivative features for automatic speech recognition. 872-875 2013 INTERSPEECH https://doi.org/10.21437/Interspeech.2013-258 conf/interspeech/2013 db/conf/interspeech/interspeech2013.html#GibsonSOGN13 Fabrizio Morbini Kartik Audhkhasi Ron Artstein Maarten Van Segbroeck Kenji Sagae Panayiotis G. Georgiou David R. Traum Shrikanth S. Narayanan A reranking approach for recognition and classification of speech input in conversational dialogue systems. 49-54 2012 SLT https://doi.org/10.1109/SLT.2012.6424196 conf/slt/2012 db/conf/slt/slt2012.html#MorbiniAASSGTN12

Maarten Van Segbroeck Hugo Van hamme Advances in Missing Feature Techniques for Robust Large-Vocabulary Continuous Speech Recognition. 123-137 2011 19 IEEE Trans. Speech Audio Process. 1 https://doi.org/10.1109/TASL.2010.2045235 db/journals/taslp/taslp19.html#SegbroeckH11

Jort F. Gemmeke Maarten Van Segbroeck Yujun Wang Bert Cranen Hugo Van hamme Automatic Speech Recognition Using Missing Data Techniques: Handling of Real-World Data. 157-185 2011 Robust Speech Recognition of Uncertain or Missing Data https://doi.org/10.1007/978-3-642-21317-5_7 books/daglib/0027836 db/books/daglib/0027836.html#GemmekeSWCh11

Maarten Van Segbroeck Hugo Van hamme Unsupervised learning of time-frequency patches as a noise-robust representation of speech. 1124-1138 2009 51 Speech Commun. 11 https://doi.org/10.1016/j.specom.2009.05.003 db/journals/speech/speech51.html#SegbroeckH09

Jort F. Gemmeke Yujun Wang Maarten Van Segbroeck Bert Cranen Hugo Van hamme Application of noise robust MDT speech recognition on the SPEECON and speechdat-car databases. 1227-1230 2009 INTERSPEECH https://doi.org/10.21437/Interspeech.2009-354 conf/interspeech/2009 db/conf/interspeech/interspeech2009.html#GemmekeWSCh09 Maarten Van Segbroeck Hugo Van hamme Applying non-negative matrix factorization on time-frequency reassignment spectra for missing data mask estimation. 2511-2514 2009 INTERSPEECH https://doi.org/10.21437/Interspeech.2009-376 conf/interspeech/2009 db/conf/interspeech/interspeech2009.html#Segbroeckh09 Maarten Van Segbroeck Hugo Van hamme Robust speech recognition using missing data techniques in the prospect domain and fuzzy masks. 4393-4396 2008 ICASSP https://doi.org/10.1109/ICASSP.2008.4518629 conf/icassp/2008 db/conf/icassp/icassp2008.html#SegbroeckH08 Maarten Van Segbroeck Hugo Van hamme Vector-quantization based mask estimation for missing data automatic speech recognition. 910-913 2007 INTERSPEECH https://doi.org/10.21437/Interspeech.2007-330 conf/interspeech/2007 db/conf/interspeech/interspeech2007.html#Segbroeckh07 Maarten Van Segbroeck Hugo Van hamme Handling convolutional noise in missing data automatic speech recognition. 2006 INTERSPEECH https://doi.org/10.21437/Interspeech.2006-642 conf/interspeech/2006 db/conf/interspeech/interspeech2006.html#Segbroeckh06 Ron Artstein David C. Atkins Kartik Audhkhasi Stefan Baack Mitchell Baker Ayah Bdeir Julie Belião Stella Biderman Matthew BlackMatthew P. Black Jillian Bommarito Daniel Bone Gurpreet Chadha Sandeep Nallan Chakravarthula Theodora Chaspari I-Fan Chen Kasia Chmielinski Erika Pelaez Coyotl Bert Cranen Jennifer Ding Jasha Droppo Peter Drotár Marzieh Fadaee Maximilian Gahntz Sriram Ganapathy Jort F. Gemmeke Panayiotis G. Georgiou Prasanta Kumar Ghosh James Gibson Tanaya Guha Suren Gunturu Jinxi Guo Rahul Gupta 0001 Lisa Gutermuth Hugo Van hamme Kyu Jeong Han Di He 0004 Mohammed Hethnawi Björn Hoffmeister Che-Wei Huang Cirenia Huerta Zac E. Imel Athanasios KatsamanisNassos Katsamanis Paul Keller Jangwon Kim Brian John King Naveen Kumar 0004 Ksenia Kutsenko Hynek Kydlícek Pierre-Carl Langlais Solana Larsen Sungbok Lee Greg Leppert EM Lewis-Jong Ming Li 0026 Greg Lindahl Shayne Longpre Angela Oduor Lungati Xuewen Luo Roland Maas Sebastian Majstorovic Nikos MalandrakisNikolaos Malandrakis Sri Harish Reddy MallidiSri Harish Mallidi Nik Marda Cullen Miller Victor Miller Fabrizio Morbini Shri NarayananShrikanth S. Narayanan Md. Nasir Tinh Nguyen Kasia Odrozek Mohamed Kamal Omar Maurizio Omologo Antonio Ortega Pavlos Papadopoulos Guilherme Penedo Alexandros Potamianos Vikram Ramanarayanan Ariya Rastrow Max Ryabinin Kenji Sagae George Saon Andreas Schwarz Kathleen Siminyu Z.-I. SkordilisZisis Iason Skordilis Aviya Skowron Andreas Stolcke Andrew Strait Mark Surman Samuel Thomas 0001 Gautam Tiwari David R. Traum Ruchir Travadi Jan Trmal Andreas Tsiartas Anna Tumadóttir Colin Vaz Jennifer Wang Yujun Wang Maurice Weber Rebecca Weiss Leandro von Werra Lee White Thomas Wolf 0008 Wei Xia Bo Xiao 0003 Zhaojun Yang Liwen You Ahmed Zaid