A two-stage active learning method enhances speech recognition accuracy with less data.
― 5 min read
Electrical Engineering and Systems Science - Audio and Speech Processing
RSSLatest Articles
Latest Articles
A device helps focus on specific voices in crowded places.
― 6 min read
A new method improves audio editing using diffusion models for precise changes.
― 5 min read
SpeechVerse bridges audio understanding and language processing for improved human-computer interaction.
― 6 min read
New dataset highlights performance gaps among demographic groups using voice assistants.
― 6 min read
This article investigates vulnerabilities in speech models and ways to enhance their security.
― 5 min read
Enhanced speech recognition for classrooms using advanced training techniques improves learning.
― 6 min read
Understanding and mitigating hallucination in AI for reliable performance.
― 7 min read
A novel approach employs graph convolutional networks for efficient music data analysis.
― 8 min read
New methods improve connections between audio clips and text descriptions.
― 5 min read
ROSVOT enhances accuracy in transcribing singing voices, even in noisy environments.
― 5 min read
New techniques enhance voice reconstruction in challenging settings using limited data.
― 7 min read
Introducing a model that generates synchronized audio and video with mixed noise levels.
― 6 min read
A new system improves robot interactions by filtering overlapping speech.
― 6 min read
This article discusses a new simple model for generating audio from images and vice versa.
― 5 min read
Denoising Language Models improve error correction in speech recognition systems using synthetic data.
― 7 min read
New model VPIDM improves clarity of speech in noisy environments.
― 6 min read
NeRAF creates synchronized sound and visuals for immersive experiences in various fields.
― 6 min read
A new method improves audio-video alignment using pre-trained models.
― 6 min read
Zipper effectively combines different data types for smarter AI models.
― 6 min read
Using deep learning to enhance acoustic emission monitoring of bolted joints.
― 7 min read
A new approach to combine singing and dance through advanced computer techniques.
― 6 min read
Learn how speech inpainting is restoring audio quality in various fields.
― 6 min read
A new system improves speech clarity in multi-speaker environments.
― 5 min read
New methods improve how machines recognize emotions in speech.
― 5 min read
Frieren model improves audio quality and sync for video.
― 6 min read
A new method generates unique sounds from text using a simple synthesizer.
― 8 min read
New method improves speech translation in noisy environments while preserving expressiveness.
― 4 min read
A new dataset enhances the study of Raga identification in Indian music.
― 5 min read
Seed-TTS creates lifelike speech from text for various applications.
― 5 min read
New method improves conversion from speech to singing using self-supervised learning.
― 7 min read
StreamSpeech improves real-time speech translation with efficiency and quality.
― 5 min read
A new model improves speech recognition using multiple decoding methods.
― 6 min read
A study on enhancing ASR for Arabic dialects using efficient model techniques.
― 5 min read
Introducing BLSP-Emo, a model that understands speech and emotions for better interactions.
― 5 min read
A recent study replicates key findings on data interpretation using sound and visuals.
― 6 min read
New model generates music using both text and visual information.
― 7 min read
A system that connects sounds with visuals, improving machine understanding.
― 6 min read
New model ARDiT improves text-to-speech synthesis and speech editing.
― 5 min read
New methods improve clarity in isolating voices from audio mixtures.
― 4 min read
Introducing SPICE, a task to improve AI interactions using contextual information.
― 7 min read