A new framework enhances voice recognition and adapts to various speech tasks.
― 4 min read
Electrical Engineering and Systems Science - Audio and Speech Processing
RSSNew methods are needed to detect advanced deepfake speech technologies.
― 5 min read
New methods boost accuracy in identifying animal sounds from limited data.
― 5 min read
New method improves virtual sound integration in AR environments.
― 6 min read
A new method aims to preserve voice privacy while allowing for effective communication.
― 4 min read
New methods improve speech recognition for low-resource languages without text.
― 4 min read
New methods enhance accuracy in speech recognition systems using phonetic understanding.
― 5 min read
This framework improves real-time animations by synchronizing speech and gestures seamlessly.
― 5 min read
New acoustic features enhance ASR systems' performance in noisy environments.
― 4 min read
A new loss function boosts audio quality by aligning phase and magnitude.
― 6 min read
A new TTS model adds emotional depth to computer-generated speech.
― 5 min read
Evaluating speech recognition models for autism diagnostic sessions.
― 6 min read
Recent methods improve audio clarity and quality using advanced models.
― 6 min read
A fresh approach improves detection of fake audio recordings.
― 5 min read
ESPnet-Codec enhances training and evaluation of neural codecs for audio and speech.
― 7 min read
Exploring methods to adapt RNNs for varying audio sample rates.
― 6 min read
New model achieves faster speech transcription without sacrificing accuracy.
― 4 min read
Discover how Matryoshka embeddings improve speaker recognition efficiency and flexibility.
― 4 min read
Introducing NanoVoice, a quick and efficient text-to-speech model for personalized audio.
― 5 min read
New model VoiceGuider improves TTS for diverse speakers.
― 6 min read
A novel method for converting voices across languages while preserving unique characteristics.
― 5 min read
New techniques improve expressive speech quality across different speakers.
― 5 min read
This article explores the role of perceptual metrics in music genre classification.
― 4 min read
A new method improves speech and audio processing across multiple tasks.
― 5 min read
A new system enhances speaker identification during discussions with multiple participants.
― 5 min read
A new framework enhances emotional expression in TTS systems.
― 5 min read
Recent findings reveal pressure sensors can be used for eavesdropping.
― 4 min read
A new algorithm improves sound event detection using self-supervised learning.
― 5 min read
Research focuses on improving methods for detecting realistic fake speech.
― 5 min read
A new method streamlines audio and video creation for better synchronization.
― 5 min read
Control audio effects using simple language descriptions for easier sound adjustments.
― 5 min read
Introducing a new model and benchmark for evaluating multi-audio tasks.
― 5 min read
A new system models emotional intensity in animated characters for enhanced realism.
― 6 min read
OpenSep automates audio separation for clearer sound experiences without manual input.
― 6 min read
PALM enhances audio recognition by optimizing prompt representation and efficiency.
― 4 min read
Explore how wire turns and gauge impact guitar pickup sound.
― 7 min read
A new method improves speech recognition for long recordings.
― 5 min read
This study analyzes how audio, video, and text work together in speech recognition.
― 7 min read
A new model improves naturalness in text-to-speech systems by analyzing pitch patterns.
― 4 min read
A new model enhances speech representation for African languages, boosting inclusivity in technology.
― 5 min read