Alert button
Picture for Herman Kamper

Herman Kamper

Alert button

Visually Grounded Speech Models have a Mutual Exclusivity Bias

Mar 20, 2024
Leanne Nortje, Dan Oneaţă, Yevgen Matusevych, Herman Kamper

Viaarxiv icon

Revisiting speech segmentation and lexicon learning with better features

Jan 31, 2024
Herman Kamper, Benjamin van Niekerk

Viaarxiv icon

Voice Conversion for Stuttered Speech, Instruments, Unseen Languages and Textually Described Voices

Oct 12, 2023
Matthew Baas, Herman Kamper

Figure 1 for Voice Conversion for Stuttered Speech, Instruments, Unseen Languages and Textually Described Voices
Figure 2 for Voice Conversion for Stuttered Speech, Instruments, Unseen Languages and Textually Described Voices
Figure 3 for Voice Conversion for Stuttered Speech, Instruments, Unseen Languages and Textually Described Voices
Figure 4 for Voice Conversion for Stuttered Speech, Instruments, Unseen Languages and Textually Described Voices
Viaarxiv icon

Rhythm Modeling for Voice Conversion

Jul 12, 2023
Benjamin van Niekerk, Marc-André Carbonneau, Herman Kamper

Figure 1 for Rhythm Modeling for Voice Conversion
Figure 2 for Rhythm Modeling for Voice Conversion
Figure 3 for Rhythm Modeling for Voice Conversion
Figure 4 for Rhythm Modeling for Voice Conversion
Viaarxiv icon

Leveraging multilingual transfer for unsupervised semantic acoustic word embeddings

Jul 05, 2023
Christiaan Jacobs, Herman Kamper

Figure 1 for Leveraging multilingual transfer for unsupervised semantic acoustic word embeddings
Figure 2 for Leveraging multilingual transfer for unsupervised semantic acoustic word embeddings
Figure 3 for Leveraging multilingual transfer for unsupervised semantic acoustic word embeddings
Figure 4 for Leveraging multilingual transfer for unsupervised semantic acoustic word embeddings
Viaarxiv icon

Disentanglement in a GAN for Unconditional Speech Synthesis

Jul 04, 2023
Matthew Baas, Herman Kamper

Figure 1 for Disentanglement in a GAN for Unconditional Speech Synthesis
Figure 2 for Disentanglement in a GAN for Unconditional Speech Synthesis
Figure 3 for Disentanglement in a GAN for Unconditional Speech Synthesis
Figure 4 for Disentanglement in a GAN for Unconditional Speech Synthesis
Viaarxiv icon

Visually grounded few-shot word learning in low-resource settings

Jun 21, 2023
Leanne Nortje, Dan Oneata, Herman Kamper

Figure 1 for Visually grounded few-shot word learning in low-resource settings
Figure 2 for Visually grounded few-shot word learning in low-resource settings
Figure 3 for Visually grounded few-shot word learning in low-resource settings
Figure 4 for Visually grounded few-shot word learning in low-resource settings
Viaarxiv icon

Towards hate speech detection in low-resource languages: Comparing ASR to acoustic word embeddings on Wolof and Swahili

Jun 01, 2023
Christiaan Jacobs, Nathanaël Carraz Rakotonirina, Everlyn Asiko Chimoto, Bruce A. Bassett, Herman Kamper

Figure 1 for Towards hate speech detection in low-resource languages: Comparing ASR to acoustic word embeddings on Wolof and Swahili
Figure 2 for Towards hate speech detection in low-resource languages: Comparing ASR to acoustic word embeddings on Wolof and Swahili
Figure 3 for Towards hate speech detection in low-resource languages: Comparing ASR to acoustic word embeddings on Wolof and Swahili
Figure 4 for Towards hate speech detection in low-resource languages: Comparing ASR to acoustic word embeddings on Wolof and Swahili
Viaarxiv icon

Voice Conversion With Just Nearest Neighbors

May 30, 2023
Matthew Baas, Benjamin van Niekerk, Herman Kamper

Figure 1 for Voice Conversion With Just Nearest Neighbors
Figure 2 for Voice Conversion With Just Nearest Neighbors
Figure 3 for Voice Conversion With Just Nearest Neighbors
Viaarxiv icon

Visually grounded few-shot word acquisition with fewer shots

May 25, 2023
Leanne Nortje, Benjamin van Niekerk, Herman Kamper

Figure 1 for Visually grounded few-shot word acquisition with fewer shots
Figure 2 for Visually grounded few-shot word acquisition with fewer shots
Figure 3 for Visually grounded few-shot word acquisition with fewer shots
Figure 4 for Visually grounded few-shot word acquisition with fewer shots
Viaarxiv icon