Speech Recognition News and Research

RSS

Tencent’s Hunyuan-Large AI Model Sets New Benchmark with 389 Billion Parameters

Hunyuan-Large, Tencent’s largest open-source Transformer-based mixture of experts (MoE) model, pushes the boundaries of AI with 389 billion parameters and 52 billion activated experts, excelling in tasks like reasoning, coding, and long-context processing. It outperforms leading models like LLama3.1, demonstrating superior scalability and efficiency.

11 Nov 2024

Lumos Enhances Multimodal AI with On-Device STR

Lumos, a multimodal AI system, integrates on-device scene text recognition (STR) to improve question answering capabilities. This innovation balances high-quality text recognition with optimized performance, advancing real-world applications for smart assistants.

28 Aug 2024

Intelligent Digital Assistants Improve Assembly Process Quality

A recent study explored the use of a large language model-based voice-enabled digital intelligent assistant in manufacturing assembly processes. It found that while the system effectively reduced cognitive load and improved product quality, it did not significantly impact lead times.

8 Aug 2024

Llama 3: Meta's New AI Model Rivals GPT-4

Meta's Llama 3, a 405B parameter transformer with a 128K token context window, matches GPT-4 in performance across various tasks. With integrated image, video, and speech capabilities, it emphasizes data quality and efficiency, though further development is needed for widespread release.

30 Jul 2024

AI and IoT Revolutionize Sports Training Analysis

Researchers have utilized AI and IoT voice devices to advance sports training feature recognition, employing sensors for real-time data transmission and analysis. This approach successfully identifies movement patterns and predicts athlete states, enhancing training effectiveness.

27 Jun 2024

Accent Classification with Deep Learning Models

A study introduces advanced deep learning models integrating DenseNet with multi-task learning and attention mechanisms for superior English accent classification. MPSA-DenseNet, the standout model, achieved remarkable accuracy, outperforming previous methods.

6 Jun 2024

Silent Speech Interface Using Graphene-Based Textile Strain Sensors and AI

Researchers introduced a groundbreaking silent speech interface (SSI) leveraging few-layer graphene (FLG) strain sensing technology and AI-based self-adaptation. Embedded into a biocompatible smart choker, the sensor achieved high accuracy and computational efficiency, revolutionizing communication in challenging environments.

15 May 2024

Smart Contact Lens for Precise Eye Tracking

This groundbreaking innovation introduces a miniature, imperceptible smart contact lens for wireless interaction, surpassing traditional eye-tracking methods. With biocompatibility confirmed through extensive testing, it heralds a new era in human-machine interaction, offering unparalleled precision and versatility.

8 May 2024

Bridging the Perception Gap: DNNs and Human Peripheral Vision

Researchers delve into the realm of object detection, comparing the performance of deep neural networks (DNNs) to human observers under simulated peripheral vision conditions. Through meticulous experimentation and dataset creation, they unveil insights into the nuances of machine and human perception, paving the way for improved alignment and applications in computer vision and artificial intelligence.

20 Mar 2024

Flash Attention Generative Adversarial Network for Enhanced Lip-to-Speech Technology

Researchers introduced the Flash Attention Generative Adversarial Network (FA-GAN) to address challenges in Chinese sentence-level lip-to-speech (LTS) synthesis. FA-GAN, incorporating joint modeling of global and local lip movements, outperformed existing models in both English and Chinese datasets, showcasing superior performance in speech quality metrics like STOI and ESTOI.

4 Mar 2024

Low-Carbon Transformation in Resource-Based Cities by Integrating ChatGPT and ABC Algorithms

Researchers propose a novel approach utilizing ChatGPT and artificial bee colony (ABC) algorithms to advance low-carbon transformation in resource-based cities. Their study demonstrates significant improvements in energy efficiency, carbon emissions reduction, and traffic congestion alleviation, highlighting the potential of these methods in promoting green development and sustainable urban planning.

15 Feb 2024

Innovative Vision Transformer for Pothole and Traffic Sign Detection in Challenging Conditions

Researchers from India, Australia, and Hungary introduce a robust model employing a cascade classifier and a vision transformer to detect potholes and traffic signs in challenging conditions on Indian roads. The algorithm, showcasing impressive accuracy and outperforming existing methods, holds promise for improving road safety, infrastructure maintenance, and integration with intelligent transport systems and autonomous vehicles

2 Feb 2024

Oracle-MNIST Dataset Unveils Challenges for ML in Ancient Chinese Character Recognition

Researchers from Beijing University introduce Oracle-MNIST, a challenging dataset of 30,222 ancient Chinese characters, providing a realistic benchmark for machine learning (ML) algorithms. The Oracle-MNIST dataset, derived from oracle-bone inscriptions of the Shang Dynasty, surpasses traditional MNIST datasets in complexity, serving as a valuable tool not only for advancing ML research but also for enhancing the study of ancient literature, archaeology, and cultural heritage preservation.

25 Jan 2024

Optical Meta-Imager Accelerates Machine Vision

Researchers present a meta-imager using metasurfaces for optical convolution, offloading computationally intensive operations into high-speed, low-power optics. The system employs angular and polarization multiplexing, achieving both positive and negative valued convolution operations simultaneously, showcasing potential in compact, lightweight, and power-efficient machine vision systems.

10 Jan 2024

Enhancing Science Education with Multimodal Large Language Models

Researchers discuss the transformative role of Multimodal Large Language Models (MLLMs) in science education. Focusing on content creation, learning support, assessment, and feedback, the study demonstrates how MLLMs provide adaptive, personalized, and multimodal learning experiences, illustrating their potential in various educational settings beyond science.

4 Jan 2024

RVTALL: Advancing Speech Recognition with Multimodal Dataset

Researchers unveil RVTALL, a groundbreaking multimodal dataset for contactless speech recognition. Integrating data from UWB and mmWave radars, depth cameras, lasers, and audio-visual sources, the dataset aids in exploring non-invasive speech analysis. The study demonstrates applications in silent speech recognition, speech enhancement, analysis, and synthesis, though it acknowledges limitations in sample size and diversity. The dataset stands as a robust tool for advancing research in speech-related technologies.

19 Dec 2023

Exploring Unique Feature Memorization in Deep Neural Networks for Image Classification

This research explores Unique Feature Memorization (UFM) in deep neural networks (DNNs) trained for image classification tasks, where networks memorize specific features occurring only once in a single sample. The study introduces methods, including the M score, to measure and identify UFM, highlighting its privacy implications and potential risks for model robustness. The findings emphasize the need for mitigation strategies to address UFM and enhance the privacy and generalization of DNNs, especially in fields like medical imaging and computer vision.

8 Dec 2023

Revolutionizing Automatic Speech Translation with Enhanced Expressivity and Multilingual Capabilities

This paper introduces a groundbreaking series of models, including SeamlessM4Tv2, SeamlessExpressive, and SeamlessStreaming, designed to advance automatic speech translation. These models excel in preserving meaning, naturalness, and expressivity, catering to diverse linguistic contexts. Safety measures, including toxicity detection, gender bias evaluation, and watermarking, underscore the commitment to ethical deployment.

5 Dec 2023

Revolutionizing Investigative Interview Training: AI-Powered Virtual Reality with Child Avatars

Researchers unveil a groundbreaking virtual reality (VR) system utilizing child avatars for immersive investigative interview training. The AI-driven prototype, featuring a lifelike 6-year-old avatar, outperforms 2D alternatives, showcasing superior realism, engagement, and training efficacy. The system's AI capabilities, including automatic performance evaluation and tailored feedback, present a promising avenue for scalable and personalized training, potentially transforming competencies in handling child abuse cases globally.

24 Nov 2023

Rainbow: An Expandable Voice User Interface for Scientific Laboratories

Researchers introduced Rainbow, an open-source Voice User Interface (VUI) designed for scientific laboratories, addressing the limitations of conventional assistants in recognizing specialized scientific vocabulary. Rainbow achieved a remarkable 91.3% speech recognition accuracy, outperforming commercial counterparts and demonstrating its potential in enhancing laboratory processes through intuitive voice control.

12 Nov 2023