Zobrazeno 1 - 10
of 29
pro vyhledávání: '"Van Rijn, Pol"'
Conversational tones -- the manners and attitudes in which speakers communicate -- are essential to effective communication. Amidst the increasing popularization of Large Language Models (LLMs) over recent years, it becomes necessary to characterize
Externí odkaz:
http://arxiv.org/abs/2406.04278
Autor:
Marjieh, Raja, van Rijn, Pol, Sucholutsky, Ilia, Lee, Harin, Griffiths, Thomas L., Jacoby, Nori
The speech-to-song illusion is a robust psychological phenomenon whereby a spoken sentence sounds increasingly more musical as it is repeated. Despite decades of research, a complete formal account of this transformation is still lacking, and some of
Externí odkaz:
http://arxiv.org/abs/2402.06992
Autor:
van Rijn, Pol, Mertes, Silvan, Janowski, Kathrin, Weitz, Katharina, Jacoby, Nori, André, Elisabeth
Speech is a natural interface for humans to interact with robots. Yet, aligning a robot's voice to its appearance is challenging due to the rich vocabulary of both modalities. Previous research has explored a few labels to describe robots and tested
Externí odkaz:
http://arxiv.org/abs/2402.05206
Autor:
van Rijn, Pol, Sun, Yue, Lee, Harin, Marjieh, Raja, Sucholutsky, Ilia, Lanzarini, Francesca, André, Elisabeth, Jacoby, Nori
Conducting experiments with diverse participants in their native languages can uncover insights into culture, cognition, and language that may not be revealed otherwise. However, conducting these experiments online makes it difficult to validate self
Externí odkaz:
http://arxiv.org/abs/2302.01614
Determining the extent to which the perceptual world can be recovered from language is a longstanding problem in philosophy and cognitive science. We show that state-of-the-art large language models can unlock new insights into this problem by provid
Externí odkaz:
http://arxiv.org/abs/2302.01308
Autor:
Marjieh, Raja, van Rijn, Pol, Sucholutsky, Ilia, Sumers, Theodore R., Lee, Harin, Griffiths, Thomas L., Jacoby, Nori
Human similarity judgments are a powerful supervision signal for machine learning applications based on techniques such as contrastive learning, information retrieval, and model alignment, but classical methods for collecting human similarity judgmen
Externí odkaz:
http://arxiv.org/abs/2206.04105
The human voice effectively communicates a range of emotions with nuanced variations in acoustics. Existing emotional speech corpora are limited in that they are either (a) highly curated to induce specific emotions with predefined categories that ma
Externí odkaz:
http://arxiv.org/abs/2205.04820
Recent advances in neural text-to-speech research have been dominated by two-stage pipelines utilizing low-level intermediate speech representation such as mel-spectrograms. However, such predetermined features are fundamentally limited, because they
Externí odkaz:
http://arxiv.org/abs/2203.16930
Autor:
van Rijn, Pol, Mertes, Silvan, Schiller, Dominik, Dura, Piotr, Siuzdak, Hubert, Harrison, Peter M. C., André, Elisabeth, Jacoby, Nori
Novel text-to-speech systems can generate entirely new voices that were not seen during training. However, it remains a difficult task to efficiently create personalized voices from a high-dimensional speaker space. In this work, we use speaker embed
Externí odkaz:
http://arxiv.org/abs/2203.15379
Autor:
van Rijn, Pol, Mertes, Silvan, Schiller, Dominik, Harrison, Peter M. C., Larrouy-Maestri, Pauline, André, Elisabeth, Jacoby, Nori
Recent TTS systems are able to generate prosodically varied and realistic speech. However, it is unclear how this prosodic variation contributes to the perception of speakers' emotional states. Here we use the recent psychological paradigm 'Gibbs Sam
Externí odkaz:
http://arxiv.org/abs/2105.01891