Multilingual generative meta-learning for spoken word classification shows small gains over monolingual models, with unique data volume mattering more than the number of languages.
Proceedings of the 34th International Conference on Neural Information Processing Systems , articleno =
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CL 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
GeMCL achieves stable 1000-class few-shot spoken word classification with 5 shots per class, comparable to finetuned HuBERT but 2000x faster adaptation using less data and time.
citing papers explorer
-
Does language matter for spoken word classification? A multilingual generative meta-learning approach
Multilingual generative meta-learning for spoken word classification shows small gains over monolingual models, with unique data volume mattering more than the number of languages.
-
Scaling few-shot spoken word classification with generative meta-continual learning
GeMCL achieves stable 1000-class few-shot spoken word classification with 5 shots per class, comparable to finetuned HuBERT but 2000x faster adaptation using less data and time.