A new method enhances ASR models for individual users using quantisation and adaptation.
― 6 min read
Computer Science - Sound
RSSNew methods enhance vocoder performance with limited audio data.
― 5 min read
A look into dysarthria, its detection, and the role of technology.
― 6 min read
Soft prompts enhance speech recognition technology for better performance in noisy environments.
― 5 min read
Research combines self-supervised learning and new measurement techniques for improved speech inversion.
― 5 min read
Researchers develop a new framework to enhance speech clarity for electrolaryngeal users.
― 5 min read
This study explores training strategies to enhance detection of fake audio.
― 5 min read
New models adapt to improve speech recognition efficiency and responsiveness.
― 5 min read
RECAP uses advanced techniques to generate accurate audio captions without retraining.
― 5 min read
A practical guide to understanding music theory through harmony and scales.
― 7 min read
A new method uses synthetic data to enhance ASR systems in unfamiliar areas.
― 6 min read
A new audio-based method estimates crowd sizes without invading personal privacy.
― 5 min read
A new approach to speech recognition enhances user interaction with flexible instructions.
― 4 min read
A robust approach to identify audio anomalies and combat voice spoofing.
― 5 min read
A new model enhances understanding of emotions during conversations.
― 5 min read
This study examines if learned speech symbols mimic word frequency patterns.
― 5 min read
Introducing a faster method for high-quality speech synthesis using diffusion models.
― 6 min read
HiFTNet offers faster, high-quality speech synthesis using efficient innovative techniques.
― 5 min read
New method transforms voices using facial features for diverse applications.
― 8 min read
AV-SUPERB evaluates audio and visual models across various tasks for better performance.
― 5 min read
A new approach enhances speaker diarization by integrating semantic data into the process.
― 5 min read
New method improves speed and efficiency in Text-to-Audio generation.
― 4 min read
Research shows improved accuracy in recognizing emotions from speech across languages.
― 4 min read
Explore how TTT enhances speech recognition by adapting to distribution shifts.
― 6 min read
Improving the way we identify sound sources using audio-visual data.
― 6 min read
A method to visualize and predict sounds in various environments using advanced technology.
― 5 min read
New methods combine audio and metadata for better language recognition.
― 5 min read
A system designed to detect voice presentation attacks enhances security in voice recognition.
― 6 min read
Enhancing Whisper's speech recognition for Vietnamese and other low-resource languages.
― 4 min read
FluentEditor improves audio editing by focusing on natural flow and consistency.
― 4 min read
Improving real-time translation through advanced segmentation techniques.
― 5 min read
Improving real-time translations through innovative methods and smart policies.
― 5 min read
Efforts to improve ASR systems for Tunisian Arabic and code-switching.
― 5 min read
Innovative methods aim to tailor music generation to user preferences.
― 6 min read
A new model improves speech separation efficiency and performance.
― 5 min read
A new approach assesses audio quality using multiple microphones in various environments.
― 5 min read
A new method enhances sound separation across different frequencies.
― 5 min read
Explore advancements in echo cancellation to enhance call quality.
― 4 min read
A new method improves music generation by adding performance context.
― 6 min read
A new approach generates audio captions using only text, improving data efficiency.
― 7 min read