Multimedia Analysis & Analytics
Developing algorithmic techniques and methodologies to drive search, discovery, retrieval, organization and consumption of multimedia content. Focusing on the analysis of three fundamental categories of multimedia: text, video and audio (including speech) to automate the generation of metadata descriptors for content. This metadata enables enhanced search and discovery techniques for multimedia queries for content discovery (recommendation services), and organization of complex data into a more understandable form (summarization, incident diarization). See select features below or use keywords to search the entire database of publications. Technical publications may also be viewed by technology area (see links, at right), with subscriptions available to specific Motorola RSS Feeds. Live RSS Feed
Technical Search Advanced
  uWave: Accelerometer-based Personalized Gesture Recognition and Its Applications (Area: Multimedia Analysis & Analytics; Type: Conference Paper; Author: Jiayang Liu)  
  March 2009  —  The proliferation of accelerometers on consumer electronics has brought an opportunity for interaction based on gestures or physical manipulation of the devices. We present uWave, an efficient ...  
 
  Software Architectures for Networked Mobile Speech Applications (Area: Multimedia Analysis & Analytics; Type: Journal Article; Author: James Ferrans)  
  March 2008  —  We examine architectures for mobile speech applications. These use speech engines for synthesizing audio output and for recognizing audio input; a key architectural decision is whether to embed these ...  
 
  Caller Tunes or Receiver Tunes? Factors Affecting Calling Experience (Area: Multimedia Analysis & Analytics; Type: Technical Report; Author: Dhaval Joshi)  
  February 2008  —  Mobile Value added services have been gaining a lot of popularity in recent past –“Caller tunes” is one such service widely used in India. Caller tunes (also known as Ring back tones) are basically ...  
 
  Symbolic Speaker Adaptation with Phone Inventory Expansion (Area: Multimedia Analysis & Analytics; Type: Conference Paper; Author: Kyung-Tak Lee)  
  April 2003  —  This paper further develops a previously proposed adaptation method for speech recognition called Symbolic Speaker Adaptation (SSA). The basic idea of SSA is to model a speaker’s pronunciation as a ...  
 
  How Does the Speaker Verification System Resist the Attack of Playback? (Area: Multimedia Analysis & Analytics; Type: Conference Paper; Author: Wei Huang)  
  September 2007  —  In this paper we designed a novel embedded speaker verification system (VoverII) to resist the attack of playback. VoverII is a two-steps text-constrained speaker verification system using machine ...