Search papers, labs, and topics across Lattice.
University of Eastern Finland
3
0
2
Multilingual SSL speech models can predict articulatory movements with surprising accuracy, outperforming monolingual models even with minimal training data.
Cross-language mismatches in acoustic-to-articulatory inversion can degrade performance by up to 20%, underscoring the limitations of current AAI models in multilingual applications.
Permutation-aware training can drastically reduce microphone bleed in music recordings, enhancing track isolation and mixing quality.
