Latest Articles for ASR

A resource-efficient approach to backdoor attacks on advanced machine learning models.

Harnessing early-exit models for efficient federated learning in ASR systems.

Denoising Language Models improve error correction in speech recognition systems using synthetic data.

New model VPIDM improves clarity of speech in noisy environments.

A study on desktop robots using natural language and visual recognition technologies.

New methods improve language model predictions under varying input conditions.

A new model improves speech recognition using multiple decoding methods.

A fresh method for testing language model safety and multilingual skills.

A new defense strategy for LLMs against backdoor attacks.

A new method combines acoustic features and confidence scores for better error correction.

This study presents a dataset and method to enhance Chinese ASR accuracy using Pinyin.

This study evaluates speech technology in low-resource languages like Tunisian Arabic.

Emilia provides a diverse dataset for improving speech generation models.

This article discusses ways to enhance numeric expression formatting in automatic transcripts.

A new model aims to improve speech translation quality through integrated systems.

AI models enhance accuracy of speech-to-text conversions.

Research enhances ASR systems using language models for better accuracy.

A method to enhance speech recognition quality in noisy environments.

A new method enhances product searches across different media formats.

SAGE-RT creates synthetic data to improve language model safety assessments.

New methods improve voice quality assessments for patients with vocal system issues.

A look at measuring accuracy in speech recognition systems with new methods.

New method enhances ASR accuracy using language models for better transcriptions.

New methods improve speech recognition in challenging multi-speaker situations.

A new method leverages speech data to improve autism assessments.

Research on modular ASR systems aims to improve performance in noisy environments.

Sortformer integrates speaker diarization and ASR for improved audio processing.

A new approach enhances ASR by focusing on specific speaker details.

An easy-to-use tool for fine-tuning speech models without complex code.

A new model helps robots follow unclear human instructions more effectively.

CADA-GAN enhances ASR systems' performance across various recording environments.

A new method improves speech interactions by integrating recognition and response processes.

A look at the Codec-SUPERB challenge results and codec performance metrics.

A project improves speech recognition for the Malasar language using Tamil resources.

Mamba enhances speech recognition with speed and accuracy, reshaping interaction with devices.

This project aims to standardize Bangla dialects for clearer communication.

A new ASR system enhances medical speech recognition for accurate patient care.

Efficiently tracks speakers in multilingual settings using automatic speech recognition.

New model improves Chinese speech recognition accuracy significantly.

Efforts to document and preserve the endangered Neo-Aramaic language.