New method improves TTS adaptation with minimal data requirements.
― 6 min read
Electrical Engineering and Systems Science - Audio and Speech Processing
RSSAn overview of explainable AI methods in automatic speech recognition.
― 6 min read
A new model improves how machines understand and respond to audio questions.
― 5 min read
Research highlights the need for improved turn-taking in TTS technology.
― 6 min read
BabySLM evaluates how well machines learn to understand speech based on children's language.
― 7 min read
A new method improves synthetic speech selection for enhanced ASR system accuracy.
― 6 min read
A new method aligns disfluent speech with text efficiently.
― 5 min read
Improving systems for silent speech recognition with new techniques.
― 5 min read
New methods enhance automatic speech recognition for rare words using context.
― 6 min read
A new method for training keyword spotting models using weak supervision in noisy environments.
― 6 min read
Methods to improve speech translation systems for underrepresented languages.
― 4 min read
MERT addresses music modeling challenges through innovative self-supervised learning techniques.
― 6 min read
A new approach enhances RNN-T performance in automatic speech recognition.
― 6 min read
AVLIT model combines sound and video for better speech clarity in noisy settings.
― 6 min read
Examining the impact of biased data in audio detection technologies.
― 6 min read
A new method enhances voice separation using multiple microphones without labeled data.
― 4 min read
A study improves speaker verification models for better identity protection.
― 6 min read
New models improve how machines respond to audio-based questions.
― 5 min read
Research aims to improve language detection in English-Mandarin conversations.
― 7 min read
New methods enhance speech synthesis for Swiss German from standard German text.
― 5 min read
Exploring methods for improved multilingual speech recognition in Indian languages.
― 6 min read
Discover how SVVAD improves voice activity detection for better speaker verification.
― 5 min read
A new method improves pronunciation feedback for language learners.
― 6 min read
A new framework evaluates how well speech models adapt to specific tasks.
― 6 min read
Research improves multilingual speech translation using semantic knowledge.
― 4 min read
HuBERT models improve speech tasks using multiple resolutions for better performance.
― 5 min read
New techniques improve accuracy in recognizing speakers and detecting imposters.
― 4 min read
A new approach enhances phase response in virtual audio effects using deep learning.
― 5 min read
SlothSpeech reveals vulnerabilities in speech recognition systems, slowing them down significantly.
― 5 min read
UnDiff enhances audio quality using innovative speech restoration techniques.
― 5 min read
Researchers examine how GSLM processes speech in noisy environments.
― 6 min read
New methods in machine learning enhance stuttering detection capabilities.
― 5 min read
EmoMix enables the creation of speech expressing mixed emotions with precise intensity.
― 5 min read
Discover the innovative Multi-Window Masked Autoencoder method for enhanced audio processing.
― 5 min read
A novel method merges audio and visual data to repair missing speech.
― 6 min read
Exploring methods for detecting hate speech in audio broadcasts of under-resourced languages.
― 4 min read
A new method restores lost high frequencies in historical recordings.
― 7 min read
A new method enhances automatic speech recognition systems for better accuracy and adaptability.
― 6 min read
A new model improves sound diffraction in virtual environments.
― 6 min read
Contextual biasing enhances ASR systems, improving accuracy in specialized tasks.
― 5 min read