Computer Vision and Pattern RecognitionDetecting Humor in Videos with FunnyNet-WA new model identifies funny moments in videos using visual, audio, and text data.
― 6 min read
Dielectric elastomers convert electrical energy into mechanical motion, offering diverse applications.
― 7 min read
ASR transcripts with errors can help identify Alzheimer's more accurately.
― 7 min read
ELLA-V enhances text-to-speech quality and control, surpassing previous models.
― 5 min read
A new approach improves animal call detection accuracy without arbitrary thresholds.
― 6 min read
A new model integrates audio and text for better speech classification.
― 6 min read
A new initiative to improve transcription technology for meetings in large rooms.
― 7 min read
New methods enhance accuracy in noisy speech recognition using large language models.
― 6 min read
Analyzing hen sounds helps improve their health and farm productivity.
― 7 min read
A method to help the visually impaired recognize sounds in mixed reality.
― 5 min read
This article discusses solutions for speech applications in languages with limited transcribed data.
― 6 min read
Researchers combine generative and discriminative methods for improved sound classification.
― 6 min read
A new model improves voice identification security and resists voice spoofing.
― 5 min read
A look at Gaussian Adaptive Attention for improved AI performance.
― 6 min read
Research shows deep learning improves our grasp of language rhythm.
― 6 min read
CoAVT integrates audio, visual, and text data for enhanced understanding.
― 7 min read
E-SHARC improves speaker identification in various audio environments.
― 6 min read
A new system generates music tailored to express happiness and sadness.
― 5 min read
A guide to understanding music similarity in generative models.
― 9 min read
A study on sound synthesis and its evaluation in controlled environments.
― 4 min read
A new method enhances accuracy in locating moving sound sources using microphone arrays.
― 6 min read
PAM offers a novel way to measure audio quality without needing reference recordings.
― 6 min read
Audio Flamingo excels in listening, conversing, and adapting to new audio tasks.
― 5 min read
A new model enhances machines' understanding of spatial audio.
― 5 min read
A new model improves speech-to-text efficiency in real-time applications.
― 6 min read
This study assesses sounds versus words in reconstructing language family trees.
― 6 min read
New model improves music creation using user feedback.
― 7 min read
Reborn offers innovative solutions for automatic speech recognition without labeled data.
― 6 min read
A new tool helps users modify sounds easily through simple text instructions.
― 8 min read
A new model merges spoken and written language for improved communication.
― 6 min read
A look at new models for natural spoken responses.
― 6 min read
A new method integrates acoustic information into language models for better speech recognition.
― 8 min read
Using music to explain cancer can enhance understanding and engagement.
― 6 min read
Learn how sound localization identifies the source of sounds using advanced techniques.
― 4 min read
A new approach to synthesize voices with improved rhythm accuracy.
― 8 min read
LLMs improve accuracy in medical transcriptions, benefiting patient care.
― 6 min read
A method for improving melody extraction across different music styles with minimal human effort.
― 8 min read
New methods enhance voice activity and overlap detection in speaker diarization.
― 6 min read
New method integrates speech signals for enhanced depression detection.
― 4 min read
This article discusses methods to create immersive sound fields using various arrangements.
― 5 min read