MEMS gyroscopes play a critical role in inertial navigation and motion control applications but typically suffer from a fundamental trade-off between measurement range and noise performance. Existing hardware-based solutions aimed at mitigating this issue introduce additional complexity, cost, and scalability challenges. Deep-learning methods primarily focus on noise reduction and typically require precisely aligned ground-truth signals, making them difficult to deploy in practical scenarios and leaving the fundamental trade-off unresolved. To address these challenges, we introduce Mixture of Experts for MEMS Gyroscopes (MoE-Gyro), a novel self-supervised framework specifically designed for simultaneous over-range signal reconstruction and noise suppression. MoE-Gyro employs two experts: an Over-Range Reconstruction Expert (ORE), featuring a Gaussian-Decay Attention mechanism for reconstructing saturated segments; and a Denoise Expert (DE), utilizing dual-branch complementary masking combined with FFT-guided augmentation for robust noise reduction. A lightweight gating module dynamically routes input segments to the appropriate expert. Furthermore, existing evaluation lack a comprehensive standard for assessing multi-dimensional signal enhancement. To bridge this gap, we introduce IMU Signal Enhancement Benchmark (ISEBench), an open-source benchmarking platform comprising the GyroPeak-100 dataset and a unified evaluation of IMU signal enhancement methods. We evaluate MoE-Gyro using our proposed ISEBench, demonstrating that our framework significantly extends the measurable range from 450 deg/s to 1500 deg/s, reduces Bias Instability by 98.4%, and achieves state-of-the-art performance, effectively addressing the long-standing trade-off in inertial sensing.

As attitude and motion sensing components, inertial sensors are widely used in various portable devices. But the severe errors of inertial sensors restrain their function, especially the trajectory recovery and semantic recognition. As a mainstream signal processing method, wavelet is hailed as the mathematical microscope of signal due to the plentiful and diverse wavelet basis functions. However, complicated noise types and application scenarios of inertial sensors make selecting wavelet basis perplexing. To this end, we propose a wavelet dynamic selection network (WDSNet), which intelligently selects the appropriate wavelet basis for variable inertial signals. In addition, existing deep learning architectures excel at extracting features from input data but neglect to learn the characteristics of target categories, which is essential to enhance the category awareness capability, thereby improving the selection of wavelet basis. Therefore, we propose a category representation mechanism (CRM), which enables the network to extract and represent category features without increasing trainable parameters. Furthermore, CRM transforms the common fully connected network into category representations, which provide closer supervision to the feature extractor than the far and trivial one-hot classification labels. We call this process of imposing interpretability on a network and using it to supervise the feature extractor the feature supervision mechanism, and its effectiveness is demonstrated experimentally and theoretically in this paper. The enhanced inertial signal can perform impracticable tasks with regard to the original signal, such as trajectory reconstruction. Both quantitative and visual results show that WDSNet outperforms the existing methods. Remarkably, WDSNet, as a weakly-supervised method, achieves the state-of-the-art performance of all the compared fully-supervised methods.

Every year, I look forward to our annual Nuance Research Conference (NRC), an opportunity for our global research organization to come together to collaborate, converse, and drive forward Nuance's vision for designing and building intelligent, conversational solutions. This year, Deep Learning and AI took center stage along with sessions focused on advancing innovation in ASR, TTS, speech signal enhancement, and language innovation (to name a few), and our discussions have primarily centered around how we at Nuance can continue to pioneer advancements that improve the dialogue between people and technology. We were fortunate to kick off the NRC with a presentation given by John Searle, a thought leader in theories of the mind and distinguished professor at the University of California, Berkeley. Prof. Searle is well known for his work on speech acts, the "Chinese room" thought experiment, and, as we found out during the course of his talk, a passionate user of Dragon dictation. Searle spoke extensively about AI and the associated problems of consciousness, arguing that a state of consciousness that is inclusive of all the feelings and sentience of a being is a prerequisite of true intelligence. So, till we find it how consciousness originates in the brain we won't be able to reproduce it in machines, and we can't claim today's AI systems to be accurate models of the human brain.

ABSTRACT We propose a spatial diffuseness feature for deep neural network (DNN)-based automatic speech recognition to improve recognition accuracy in reverberant and noisy environments. The feature is computed in real-time from multiple microphone signals without requiring knowledge or estimation of the direction of arrival, and represents the relative amount of diffuse noise in each time and frequency bin. It is shown that using the diffuseness feature as an additional input to a DNN-based acoustic model leads to a reduced word error rate for the REVERB challenge corpus, both compared to logmelspec features extracted from noisy signals, and features enhanced by spectral subtraction. Index Terms-- Speech Recognition, Reverberation, Diffuse Noise, Deep Neural Networks 1. INTRODUCTION In automatic speech recognizers (ASR) based on Gaussian Mixture Models and Hidden Markov Models (GMM-HMM), a wide variety of transformations and feature extraction steps is currently being employed with the aim of extracting and normalizing the information contained in the time-domain input signal as efficiently as possible. Recently, with the development of effective training methods for acoustic models based on multiple-layer neural networks, which are often summarized under the term "deep neural networks" (DNN) [1], it has become possible for the acoustic model to learn relationships between features and phonemes to a higher degree than it is possible with manually implemented feature transformation steps.