Upload
darlene-curtis
View
222
Download
0
Tags:
Embed Size (px)
Citation preview
Speaking to Computers
Alex AceroManager, Speech Research GroupMicrosoft Research
[email protected] Feb 14th 2003
Talk Outline
Role of speech technology in devices
Telephony Smartphones and PDAs Multimodality in User Interface
The Promise of Speech Technology
HighHigh
InternetInternetTVTV
PhonePhone
PDAPDA
Ease of text input (keyboard/pen)Ease of text input (keyboard/pen)
Ease Ease of GUIof GUI
(screen/(screen/Pointer)Pointer)
LowLow HighHigh
PCPC
TabletTabletPCPC
ScreenScreenPhonePhoneScreenScreenPhonePhone
PDAPDA
TabletTabletPCPC
CarCarCarCar
InternetInternetTVTV
Role of Speech in Different Devices
PhonePhone
PCPC
ScreenScreenPhonePhone
PDAPDA
TabletTabletPCPC
CarCar
InternetInternetTVTV
A Roadmap for Speech
Ease of text input (keyboard/pen)Ease of text input (keyboard/pen)
Ease Ease of GUIof GUI
(screen/(screen/Pointer)Pointer)
HighHigh
HighHighLowLow
Speech-Only Speech-Only TelephonyTelephony
DictationDictation
Multimodal Multimodal Command/ControlCommand/Control
Speech Technology
Meeting / Voicemail Transcription
Market Opportunity
Mobile Devices / Cars
Telephony / Call Center
Accessibility
Desktop Dictation
Desktop Command & Control
Technology Readiness
Customer Need
PoorAlternative
The Business Value of Speech for Call Centers
Customer Focus
Less Time/Call
Efficient Agents
Less Time in Queue
Increased System Usage
Customer Retention
$5/call to $.20/call
Reduced Call Time
Fewer Agents
New Revenue Opportunities
Up-Sell/Cross-Sell
Amtrak61% Increase in Satisfaction
75% Increase in Automation Rate
90% Increase in Ticket Sales
Thrifty Car Rental40% increase in CSR productivity $1 million first year savings
Merrill LynchAutomation rates from 82% to 90%
First Year Savings $6.3M
Call Center Examples
The Business Value of Speech for Operators
0
5000
10000
15000
20000
25000
30000
35000
2000 2001 2002 2003 2004 2005 2006 2007
Data Revenue
Voice RevenueRevenueIn US$M
The mobile operators need to make money from value-added services!
If you still doubt speech is goodfor the call center….
Why Speech at Microsoft?
Natural UI, or the combination of speech recognition, natural language understanding, automatic learning... Those are the key technologies that will have the most impact over the next 15 years.
Bill Gates, Microsoft Chairman
Microsoft Speech Server & SDK
Visual Studio + ASP.NET + SALT
Multiple Devices
Call center + multimodal solution
Unifies web & call center
Reduces TCO
Speech in Mobile Devices
Microsoft Smartphone & PocketPC Phones• Rich Client• 3% to 16% of WW mobile phone market
Smartphones• Thin Client• 11% to 25% of WW mobile phone market
Cellular Phones• No Client• 86% to 59% of WW mobile phone market
SOURCE: Gartner, IDC, Microsoft
2004 2007
Thin Client Devices Over Voice Channel
Web ServerMS Speech Server
PSTN
SMS Messages
Voic
e O
nly
Ap
ps
GrammarsGrammars
PromptsPrompts
ASP.NETDialogs
ASP.NETDialogs
Speech EngineServices
Speech EngineServices
Telephony AppServices
Telephony AppServices
Rich Client Devices Over Data Channel
Web ServerMS Speech Server
SMS Push for Brower Launch
Microsoft Voice Command
Pocket PC voice-enabled applications: Voice Dialer, Contacts, Calendar, Media
Player No connectivity necessary (100%
embedded) No training needed, (speaker-
independent) Continuous speech recognition
“Call John at home”
Multimodal Interactive Pad (MIPAD)
Multimodal Map
Current Speech User Interfaces
Need improved Speech user interfaces Even no-errors and fast processing not sufficient But errors occur: better error correction needed
Social issues: Microphones can’t tether user Users more comfortable talking to phones, cars. Talking to computers not likely in meetings or
cubicles
The Future of Natural User Interfaces
End User End User NeedsNeeds
Technology, Technology, ResearchResearch
Software ScenariosSoftware Scenarios
Bridging The Gap
Thank You!
http://research.microsoft.com/srg