Zobrazeno 1 - 10
of 27
pro vyhledávání: '"Bartelds, Martijn"'
Autor:
Shi, Jiatong, Wang, Shih-Heng, Chen, William, Bartelds, Martijn, Kumar, Vanya Bannihatti, Tian, Jinchuan, Chang, Xuankai, Jurafsky, Dan, Livescu, Karen, Lee, Hung-yi, Watanabe, Shinji
ML-SUPERB evaluates self-supervised learning (SSL) models on the tasks of language identification and automatic speech recognition (ASR). This benchmark treats the models as feature extractors and uses a single shallow downstream model, which can be
Externí odkaz:
http://arxiv.org/abs/2406.08641
The performance of automatic speech recognition (ASR) systems has advanced substantially in recent years, particularly for languages for which a large amount of transcribed speech is available. Unfortunately, for low-resource languages, such as minor
Externí odkaz:
http://arxiv.org/abs/2305.10951
Autor:
San, Nay, Bartelds, Martijn, Billings, Blaine, de Falco, Ella, Feriza, Hendi, Safri, Johan, Sahrozi, Wawan, Foley, Ben, McDonnell, Bradley, Jurafsky, Dan
Recent research using pre-trained transformer models suggests that just 10 minutes of transcribed speech may be enough to fine-tune such a model for automatic speech recognition (ASR) -- at least if we can also leverage vast amounts of text data (803
Externí odkaz:
http://arxiv.org/abs/2302.04975
Autor:
Bartelds, Martijn, Wieling, Martijn
Deep acoustic models represent linguistic information based on massive amounts of data. Unfortunately, for regional languages and dialects such resources are mostly not available. However, deep acoustic models might have learned linguistic informatio
Externí odkaz:
http://arxiv.org/abs/2205.02694
Autor:
San, Nay, Bartelds, Martijn, Ògúnrèmí, Tolúlopé, Mount, Alison, Thompson, Ruben, Higgins, Michael, Barker, Roy, Simpson, Jane, Jurafsky, Dan
Many archival recordings of speech from endangered languages remain unannotated and inaccessible to community members and language learning programs. One bottleneck is the time-intensive nature of annotation. An even narrower bottleneck occurs for re
Externí odkaz:
http://arxiv.org/abs/2204.07272
Autor:
Brouwer, Jelle, Buurke, Raoul, van den Berg, Floor, Knooihuizen, Remco, Loerts, Hanneke, Bartelds, Martijn, Wieling, Martijn, Keijzer, Merel
Publikováno v:
In Ampersand June 2024 12
Publikováno v:
Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021
For many (minority) languages, the resources needed to train large models are not available. We investigate the performance of zero-shot transfer learning with as little data as possible, and the influence of language similarity in this process. We r
Externí odkaz:
http://arxiv.org/abs/2105.02855
Autor:
San, Nay, Bartelds, Martijn, Browne, Mitchell, Clifford, Lily, Gibson, Fiona, Mansfield, John, Nash, David, Simpson, Jane, Turpin, Myfany, Vollmer, Maria, Wilmoth, Sasha, Jurafsky, Dan
Pre-trained speech representations like wav2vec 2.0 are a powerful tool for automatic speech recognition (ASR). Yet many endangered languages lack sufficient data for pre-training such models, or are predominantly oral vernaculars without a standardi
Externí odkaz:
http://arxiv.org/abs/2103.14583
Autor:
Bartelds, Martijn, de Vries, Wietse, Sanal, Faraz, Richter, Caitlin, Liberman, Mark, Wieling, Martijn
Variation in speech is often quantified by comparing phonetic transcriptions of the same utterance. However, manually transcribing speech is time-consuming and error prone. As an alternative, therefore, we investigate the extraction of acoustic embed
Externí odkaz:
http://arxiv.org/abs/2011.12649
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.