| Conversational Technology
Conversational technology allows people to get information, conduct transactions, and be entertained, simply by speaking to a computer. Based on the integration of speech recognition, natural language processing and dialog understanding technologies, conversational systems are rapidly becoming an important component of solutions for customer care, mobile access to data and many other applications. There are now hundreds of currently deployed systems that use these speech technologies. These systems save money and time, attract new customers, and improve accessibility for their users on a day in and day out basis for customers all over the world.
As exciting as these systems are, the technologies they are based on are both highly complex and rapidly evolving. You need the best and most up to date information to make informed decisions on how they can be used to address your organization’s requirements. Conversational Technologies provides the reports, analysis, training, and design services that you need to understand these technologies and the solutions that use them.
News
Natural Language Processing Report
February 26, 2008: A new report, "Natural Language Processing Tools" is now available from Conversational Technologies. This report describes the current state of the art in current natural language processing tools and systems. It focuses on commercial applications, but also includes information on research systems. Please contact us for further information.
Voice Search Conference February 26, 2008:Deborah Dahl will be talking on "New W3C Standards for Speech and Multimodal Applications" at the Voice Search Conference, March 10-12 in San Diego. This presentation will discuss some of the newer W3C standards in the voice and multimodal interaction area and how they can be applied to create compelling voice search applications.
EMMA
December 11, 2007: The W3C has published EMMA as a Candidate Recommendation. EMMA is a standard format for representing inputs from users using speech, text, pointing and a wide variety of other modalities. It will be very useful in promoting interoperability among components that accept user input such as speech recognizers. When a specification is published as a Candidate Recommendation, it means that the specification is essentially technically complete. The next step is for organizations who have implemented EMMA to submit reports on their EMMA implementations in order to verify that EMMA's features are both implementable and useful. Any organization can submit an implementation report. Reports are due April 14, 2008. See the announcement and the EMMA spec for more information.
Workshop on W3C's Multimodal Architecture and Interfaces
The W3C held a workshop at Keio University in Fujisawa, Japan, November 16-17, 2007, to discuss the W3C Multimodal Architecture. Please see the summary for details.
Presentations at SpeechTEK Fall 2007, August 20-23
1. Tutorial on Natural Language Processing: This tutorial introduces natural language processing and its role in speech applications. Topics:
-- What natural language is
-- Techniques for processing natural language
-- Emerging standards and research and their roles in future applications. We used a Natural Language Workbench demo for part of the tutorial. This demo is now available as open source software. It includes simple tools for experimenting with grammar-based and statistical speech processing as well as an implementation of EMMA. Download the natural language workbench.
You can also download the tutorial from the SpeechTek website.
2. "Using speech recognition in speech therapy: A multimodal application for users with aphasia" A case study of a multimodal application which integrates speech and GUI to provide speech therapy for individuals with aphasia, a loss of language ability resulting from brain injury.
You can download the presentation from the SpeechTek website.
Recent Publications and Presentations
Standards for Multimodal Applications VoiceXML Review, volume 7, issue 1, April-May, 2007.
10 Innovative Speech Applications Presented at the Spring 2007 SpeechTEK/AVIOS Conference, San Francisco.
Point/Counter Point on Personas Speech Technology Magazine, January-February, 2006, Vol. 11, no. 1, pp. 18-21.
Guide to Speech Standards Speech Technology Magazine, March-April, 2005, Vol. 10, no. 2, pp. 16-23.
W3C Multimodal Interaction Working Group Activities. VoiceXML Review, January-February, 2005.
W3C standards for automating customer support: Multimodal Dialogs. Service Automation Expo February 21, 2005, San Francisco
W3C Standards for Multimodal Interaction. Speechtek-AVIOS Spring Conference February 23, 2005, San Francisco
Deborah Dahl, Editor. Practical Spoken Dialog Systems. Springer-Verlag, 2005.
This book is a collection of 10 papers describing the life cycle of spoken dialog systems from requirements definition through deployment, with an emphasis on the practical considerations that are needed to make these applications successful. It also offers a glimpse into the future by including two papers on systems that are still under development in research laboratories. Some of the specific topics treated are error handling, speaker authentication, tool selection, and the psychological bases of human-computer communication.
Our Services
Getting started with conversational systems and technologies
If you're considering speech-enabling a web-based application or a touchtone-based telephone application, or developing a brand-new speech application, whether on the desktop or on the web, we can help you
- Understand the state of the art in speech technologies and the speech industry through custom reports and presentations
- Understand your application’s requirements with respect to the current state of the art in speech and dialog technologies
- Develop a plan that will result in a high-quality deployment in a predictable amount of time
- Develop rapid prototypes of prospective applications to get a hands-on feel for what the application will really be like
- Develop realistic requirements that will result in quality proposals
- Evaluate proposals for risk, realism, and cost-effectiveness
- Recommend, evaluate, and design development tools for speech applications
We have full and half-day programs available to introduce you to speech technology and how it can benefit your business.
Standards
If you’re starting to get familiar with speech technology you will encounter specifications such as VoiceXML. We can help you learn more about speech-related standards including VoiceXML, the Speech Recognition Grammar Format (SRGS) and the Speech Synthesis Markup Language (SSML) and EMMA.
Language technology evaluations
Based on our many years of experience in the natural language and speech industry, we can help you understand the potential of new language and speech related companies by assessing both their basic underlying technologies as well as their prospects for becoming commercially practical .
Professional Memberships
Applied Voice Input/Output Society (AVIOS)
Association for Computational Linguistics
Association for Computing Machinery
Independent Computer Consultants of America
Linguistic Society of America
Midwest Speech Technology Association Conversational Technologies is a participant in the Midwest Speech Technology Association Member Web Ring |