Researchers are developing synthetic voice data to protect privacy in voice recognition.
― 5 min read
Computer Science - Sound
RSSLatest Articles
Latest Articles
A look at how speech quality is tested using crowdsourcing.
― 5 min read
Advanced techniques for ensuring audio authenticity in the age of voice cloning.
― 5 min read
A new method trains audio captioning systems using only text descriptions.
― 6 min read
A guide to crafting clear and effective academic papers.
― 3 min read
Erie simplifies turning data into sound for better accessibility.
― 6 min read
Examining the risks of backdoor attacks on speaker verification systems.
― 6 min read
A new method enhances audio-visual segmentation without detailed labels.
― 5 min read
PIAVE helps machines extract voices clearly, even when speakers turn their heads.
― 6 min read
Libriheavy offers 50,000 hours of spoken English to boost speech recognition technology.
― 5 min read
AV2Wav enhances speech quality using audio and visual cues.
― 5 min read
A fresh method for machines to alter speech emotions naturally.
― 5 min read
New methods are being developed to identify deepfake singing voices in the music industry.
― 6 min read
Core-set selection improves text-to-speech models by focusing on diverse data.
― 5 min read
New models are transforming how we analyze emotions in speech.
― 6 min read
A new method uses ultrasound to recognize actions while protecting privacy.
― 5 min read
Introducing a flexible framework to enhance voice privacy research.
― 7 min read
CiwaGAN combines control of speech movements and information sharing for better speech learning.
― 6 min read
A framework that blends verbal and non-verbal cues for better language learning.
― 5 min read
A new method simplifies understanding of speech classification models.
― 6 min read
A new system enhances pronunciation skills by considering first language influences.
― 5 min read
Discover how quantum tools change music creation and performance.
― 6 min read
New method improves emotion preservation in voice conversion processes.
― 6 min read
New method preserves emotional tone in voice conversion for better human-computer interaction.
― 5 min read
New systems improve translation from text to spoken language without intermediates.
― 4 min read
Researchers enhance heart sound classification accuracy using codec data augmentation methods.
― 5 min read
Research reveals emotional speech impacts model performance in speech separation tasks.
― 6 min read
M-AUDIODEC compresses multi-channel audio while retaining speaker position and quality.
― 6 min read
New methods in S2ST improve translation quality while maintaining speaker identity.
― 5 min read
A novel system enhances spatial audio compression for clearer sound and efficiency.
― 4 min read
A new system that connects music and language for better understanding.
― 6 min read
Research reveals new models to enhance voice clarity in smart earbuds.
― 5 min read
Using extra information boosts our ability to identify bird calls.
― 5 min read
A new approach enhances audio generation by aligning audio with text descriptions.
― 5 min read
Researchers work to improve online speech recognition using structured state-space models.
― 5 min read
A new system enhances meeting experiences by identifying speakers in real-time.
― 4 min read
New methods are improving our ability to detect fake speech effectively.
― 6 min read
A method for voice conversion improving privacy and speech quality.
― 7 min read
New methods enhance ability to distinguish fake audio from real.
― 6 min read
A method improves detection of synthetic voices and identifies their creators.
― 5 min read
New methods improve tiny models for better speech enhancement using less resources.
― 5 min read