As a typical active noise control algorithm, Filtered-x Least Mean Square (FxLMS) is widely used in the field of audio denoising. In this study, an denoising coprocessor based on Retrenched Injunction System Computer-V (RISC-V), custom instruction set extension was designed and software hardware co-design adopted; traditional pure implementation, accelerator optimization design carried out, con...

In recent years podcasting has been in great demand as a recreation and a learning tool. In this paper we describe the design and implementation of a system for automatically converting documents to structured audio. Our prototype is a Web-based service for preparing structured audio material to download on portable mp3 players. The on-line service is especially designed to aid users with speci...

In this thesis, I propose a framework for designing 3D-based audio-only games in which all navigation is based on perceiving the 3D-audio, as opposed to relying on other navigational aids or imagining the audio as being spatial, where additional sounds may be added later on in the development process. To test the framework, a game named Sound Hunter was developed in an iterative process togethe...

By the integration of high speed computer networks and interactive multi-media workstations, the IRI (Interactive Remote Instruction) System creates a geographically dispersed virtual classroom. Each student participates using a personal workstation which can be used to view multi-media class presentations, to make presentations, to take notes in a multi-media notebook, and to interact via audi...

The Motorola DSP56001 is a state-of-the-art digital signal processing chip that can execute (with the clock at 25 MHz) 12.5 million instructions per second and, in a single instruction, can perform a 24 by 24-bit multiply, a 48 plus 56-bit addition, two parallel data moves, an instruction fetch and decode, and two general index updates. The 24-bit data paths and computational architecture make ...

In this paper we present example applications using a deadline instruction. The deadline instruction brings cycle accurate timing information into the application code. We have implemented the mechanism in a time-predictable Java chipmultiprocessor. As a proof of the accuracy that can be gained, a digital to analog conversion of audio signals is implemented completely in software. Furthermore, ...

Current audio player interfaces generally provide brief information such as title and duration time and support basic control functions including play, pause, rewind, and forward. These features alone are not sufficient for certain user tasks, such as quickly finding a previously-visited location or browsing the main topics covered before listening. We present VisPod, a novel visual audio playe...

Noise-robust Automatic Speech Recognition (ASR) is essential for robots which are expected to communicate with humans in a daily environment. In such an environment, Voice Activity Detection (VAD) strongly affects the performance of ASR because there are many acoustically and visually noises. In this paper, we improved Audio-Visual VAD for our two-layered audio visual integration framework for ...