Interactive voice response
|This article needs additional citations for verification. (March 2012)|
Interactive voice response (IVR) is a technology that allows a computer to interact with humans through the use of voice and DTMF tones input via keypad.
In telecommunications, IVR allows customers to interact with a company’s host system via a telephone keypad or by speech recognition, after which they can service their own inquiries by following the IVR dialogue. IVR systems can respond with prerecorded or dynamically generated audio to further direct users on how to proceed. IVR applications can be used to control almost any function where the interface can be broken down into a series of simple interactions. IVR systems deployed in the network are sized to handle large call volumes.
It is common in industries that have recently entered the telecommunications industry to refer to an automated attendant as an IVR. The terms, however, are distinct and mean different things to traditional telecommunications professionals [clarification needed (how do they differ?)], whereas emerging telephony and VoIP professionals often use the term IVR as a catch-all to signify any kind of telephony menu, even a basic automated attendant. The term voice response unit (VRU), is sometimes used as well.
- 1 History
- 2 Typical uses
- 3 Technologies used
- 4 Outsourcing vs. contact center automation
- 5 IP
- 6 See also
- 7 References
- 8 External links
Research in speech technology predated the advent of digital computers. It began with a speech synthesis project at Bell Labs in 1936 that resulted in a device called "the Voder" which was demonstrated at the 1939 World's Fair. A link between speech and mathematics resulted in a breakthrough in the early 1970s. Leonard E. Baum, and Lloyd R. Welch, invented an approach to recognition based on a statistical concept called the Hidden Markov Model. In 1961, Bell System developed a new tone dialing methodology. Bell unveiled the first telephone that could dial area codes using DTMF technology at the Seattle World Fair in 1962. DTMF telephones enabled the use of in-band signaling, i.e., they transmit audible tones in the same 300 Hz to 3.4 kHz range occupied by the human voice. The blueprint for IVR was born.
Despite the increase in deployment of IVR technology in the 1970s to automate tasks in call centers, the technology was still complex and expensive. Early voice response systems were DSP technology based, and were limited to small vocabularies. However, in the early 1980s a first mainstream market competitor emerged when Leon Ferber (Perception Technology) realized that hard drive technology (read/write random-access to digitized voice data) had finally reached a cost effective price point. A system could store digitized speech on disk, play the appropriate spoken message, and process the human's DTMF response.
As call centers began to migrate to multimedia in the late 1990s, companies started to invest in computer telephony integration (CTI) with IVR systems. IVR became vital for call centers deploying universal queuing and routing solutions and acted as an agent which collected customer data to enable intelligent routing decisions.
With improvements in technology, systems could use speaker-independent voice recognition of a limited vocabulary instead of requiring the person to use DTMF signaling.
In the subsequent decade, voice response started to become more common and cheaper to deploy. This was due to increased CPU power and the migration of speech applications from proprietary code to the VXML standard.
IVR technology is also being introduced into automobile systems for hands-free operation. Current deployment in automobiles revolves around satellite navigation, audio and mobile phone systems.
IVR systems are typically intended to service high call volumes, reduce cost and improve the customer experience. Examples of typical IVR applications are telephone banking, televoting, and credit card services. Companies also use IVR services to extend their business hours to 24/7 operation. The use of IVR and voice automation allows callers' queries to be resolved without the need for queueing and incurring the cost of a live agent. If callers do not find the information they need or require further assistance, their calls are often transferred to an agent. This makes for a more efficient system in which agents have more time to deal with complex interactions. The agents do not deal with basic inquiries that require yes/no responses or obtaining customer details.
Call centers use IVR systems to identify and segment callers. The ability to identify customers allows services to be tailored according to the customer profile. The caller can be given the option to wait in the queue, choose an automated service, or request a callback. The system may obtain caller line identification (CLI) data from the network to help identify or authenticate the caller. Additional caller authentication data could include account number, personal information, password and biometrics (such as voice print).
When an IVR system answers multiple phone numbers the use of DNIS ensures that the correct application and language is executed. A single large IVR system can handle calls for thousands of applications, each with its own phone numbers and script.
IVR also enables customer prioritization. In a system wherein individual customers may have a different status the service will automatically prioritize the individual's call and move customers to the front of a specific queue. Prioritization could also be based on the DNIS and call reason.
Smaller companies and start-ups can also use an IVR system to make their business appear larger than it is. For example, a caller never needs to know that their sales and support calls are routed to the same person.
In addition to interacting with customer information systems and databases, IVRs will also log call detail information into its own database for auditing, performance report, and future IVR system enhancements.
CTI allows a contact center or organization to gather information about the caller as a means of directing the inquiry to the appropriate agent. CTI can transfer relevant information about the individual customer and the IVR dialog from the IVR to the agent desktop using a screen-pop, making for a more effective and efficient service.
IVR may be used by survey organizations to ask more sensitive questions where the investigators are concerned that a respondent might feel less comfortable providing these answers to a human interlocutor (such as questions about drug use or sexual behavior). In some cases an IVR system can be used in the same survey in conjunction with a human interviewer. For example, during the survey the interviewer might inform the respondent that for the next series of questions they will be sent to an IVR system to continue or complete the interview.
Voice-activated dialing (VAD) IVR systems are used to automate routine enquiries to switchboard or PABX (Private Automatic Branch eXchange) operators, and are used in many hospitals and large businesses to reduce the caller waiting time. An additional function is the ability to allow external callers to page staff and transfer the inbound call to the paged person.
Entertainment and information
Some of the largest installed IVR platforms are used for televoting on television game shows, such as Pop Idol and Big Brother, which can generate enormous call spikes. Often, the network provider will have to deploy call gapping in the PSTN to prevent network overload.
IVR systems allow callers to obtain data relatively anonymously. Hospitals and clinics have used IVR systems to allow callers to receive anonymous access to test results. This is information that could easily be handled by a person but the IVR system is used to preserve privacy and avoid potential embarrassment of sensitive information or test results. Users are given a passcode to access their results.
IVR systems are used by pharmaceutical companies and contract research organizations to conduct clinical trials and manage the large volumes of data generated. The caller will respond to questions in their preferred language and their responses will be logged into a database and possibly recorded at the same time to confirm authenticity. Applications include patient randomization and drug supply management. They are also used in recording patient diaries and questionnaires.
IVR systems can be used for outbound calls, as IVR systems are more intelligent than many predictive dialer systems, and can use call progress detection to recognize different line conditions as follows:
- Answer (the IVR can tell the customer who is calling and ask them to wait for an agent)
- Answered by voice mail or answering machine (in these circumstances the IVR system can leave a message)
- Fax tone (the IVR can leave a TIFF image fax message)
- Divert messages (the IVR will abandon the call)
- No answer
Other common IVR services include:
- Mobile — pay-as-you-go account funding; registration; mobile purchases, such as ring tones and logos
- Banking — balance, payments, transfers, transaction history
- Retail & entertainment — orders, bookings, credit & debit card payments
- Utilities — meter readings; account look-up, history and balance, payment, consumption history
- Travel — ticket booking, flight information, check-in
- Weather forecasts, water, road and ice conditions
Two main varieties of speech recognition are used in IVR: that based upon predefined grammars (used in "directed" dialogues), and that based on statistically trained language models (used in "natural language" dialogues). Directed dialogues prompt the caller with specific questions or options. Natural language dialogues employ open questions (e.g. "How can I help you?"), are more conversational, and can interpret free-form responses.
Other technologies include using text-to-speech (TTS) to speak complex and dynamic information, such as e-mails, news reports or weather information. TTS is computer generated synthesized speech that is no longer the robotic voice traditionally associated with computers. Real voices create the speech in fragments that are spliced together (concatenated) and smoothed before being played to the caller.
An IVR can be deployed in several ways:
- Equipment installed on the customer premises
- Equipment installed in the PSTN (public switched telephone network)
- Application service provider (ASP) / hosted IVR
IVR can be used to provide a more sophisticated voice mail experience to the caller. For example, the IVR could ask if the caller wishes to hear, edit, forward or remove a message.
An automatic call distributor (ACD) is often the first point of contact when calling many larger businesses. An ACD uses digital storage devices to play greetings or announcements, but typically routes a caller without prompting for input. An IVR can play announcements and request an input from the caller. This information can be used to profile the caller and route the call to an agent with a particular skill set. (A skill set is a function applied to a group of call-center agents with a particular skill.)
Interactive voice response can be used to front-end a call center operation by identifying the needs of the caller. Information can be obtained from the caller such as an account number. Answers to simple questions such as account balances or pre-recorded information can be provided without operator intervention. Account numbers from the IVR are often compared to caller ID data for security reasons and additional IVR responses are required if the caller ID does not match the account record.
IVR call flows are created in a variety of ways. A traditional IVR depended upon proprietary programming or scripting languages, whereas modern IVR applications are generated in a similar way to Web pages, using standards such as VoiceXML, CCXML, SRGS and SSML. The ability to use XML-driven applications allows a web server to act as the application server, freeing the IVR developer to focus on the call flow. It was widely believed[who?] that developers would no longer require specialized programming skills; however, this has been proven to be misguided as IVR applications need to understand the human reaction to the application dialog.
Higher level IVR development tools are available to further simplify the application development process. A call flow diagram can be drawn with a GUI tool and the presentation layer (typically VoiceXML) can be automatically generated. In addition, these tools normally provide extension mechanisms for software integration, such as an HTTP interface to a web site and a Java interface for connecting to a database.
In telecommunications, an audio response unit (ARU) is a device that provides synthesized voice responses to DTMF keypresses by processing calls based on (a) the call-originator input, (b) information received from a database, and (c) information in the incoming call, such as the time of day.
ARUs increase the number of information calls handled and provide consistent quality in information retrieval.
Outsourcing vs. contact center automation
Contact centers can be expensive to run, and are often seen as cost centers; however, the ability to up-sell services and products to customers can offset operational expenditure, and effectively reduce the average cost per call handled.
Methods of reducing contact center running costs include outsourcing and automation. Outsourcing to other countries can reduce operational expenditure by as much as 30%, however, differences in culture and language can prove problematic for customers, whose dissatisfaction can lead to customer complaints and loss of business. Also, it is more difficult to up-sell to customers from foreign contact centers.
Automation in a contact center can also reduce operational expenditure by around 30% through the introduction of technologies such as customer profiling, CTI, and IVR using speech recognition. The use of automation in the contact center promotes efficiency, allowing contact centers to be located in the country from which the call is originated. Customer satisfaction can be monitored by the use of customer survey applications. The information from survey applications can be used to improve customer service.
The increased usage of IP in telecommunications has affected how IVR is used.
The introduction of Session Initiation Protocol (SIP) means that point-to-point communications are no longer restricted to voice calls but can now be extended to multimedia technologies such as video. IVR manufacturers have extended their systems into IVVR (interactive voice and video response), especially for the mobile phone networks. The use of video gives IVR systems the ability to implement multimodal interaction with the caller.
The introduction of full-duplex video IVR in the future will allow systems the ability to read emotions and facial expressions. It may also be used to identify the caller, using technology such as Iris scan or other biometric means. Recordings of the caller may be stored to monitor certain transactions, and can be used to reduce identity fraud.
Unified communications in the SIP contact center
With the introduction of SIP contact centers, traditional barriers to automation are breaking down. Call control in a SIP contact center can be implemented by CCXML scripting, which is an adjunct to the VXML language used to generate modern IVR dialogues. As calls are queued in the SIP contact center, the IVR system can provide treatment or automation, wait for a fixed period, or play music. Inbound calls to a SIP contact center must be queued or terminated against a SIP end point; SIP IVR systems can be used to replace agents directly by the use of applications deployed using BBUA (back-to-back user agents).
Interactive messaging response (IMR)
As communications have migrated to multimedia so has automation. The introduction of instant messaging (IM) in contact centers is starting to take off. Agents can handle up to 6 different IM conversations at the same time and so agent productivity is increasing. IVR technology is being used to automate IM conversations using existing natural language processing software. This is different from email handling as email automated response is typically based on key word spotting. IM conversations are different from email as IM is conversational. The use of text messaging abbreviations and smilies requires different grammars to those currently used for speech recognition. IM is also starting to replace text messaging on multimedia mobile handsets and is expected to become more widely used.
Hosted vs. on-premises IVR
With the introduction of web services into the contact center, host integration has been simplified, allowing IVR applications to be hosted remotely from the contact center. This has meant hosted IVR applications using speech are now available to smaller contact centers across the globe and has led to an expansion of ASP (application service providers).
IVR applications can also be hosted in the public network, without contact center integration. Services include public announcement messages and message services for small business. It is also possible to deploy two-prong IVR services where the initial IVR application is used to route the call to the appropriate contact center. This can be used to balance loading across multiple contact centers or provide business continuity in the event of a system outage.
IVR is sometimes criticized as being unhelpful and difficult to use due to poor design and lack of appreciation of the caller's needs. Some callers object to providing voice response to an automated system and prefer speaking with a human respondent.
Companies have also been criticized for using IVR to reduce operational costs but not offering similar services using agents. Such methods tend to frustrate customers who feel that their right to speak to an agent is being restricted. Examples of services criticized in this way include debt recovery and giveaways (concert tickets, satellite/cable receivers, etc.).
- Voice-based marketing automation
- Automatic call distributor
- Dual-tone multi-frequency (DTMF)
- Automated attendant
- Call whisper
- Speech synthesis
- Automatic number identification
- Speech recognition
- Natural language
- Dialog system
- Electronic patient-reported outcome
- The Voder
- Voice User Interface
- Khasnabish, Bhumip (2003-05-30). Implementing Voice Over IP. Lexington, Massachusetts, USA: John Wiley & Sons, Inc. p. 203. ISBN 9780471216667. Retrieved 21 March 2012.
- Lam MY, Lee H, Bright R, Korzenik JR, Sands BE (2009). "Validation of interactive voice response system administration of the Short Inflammatory Bowel Disease Questionnaire". Inflamm.Bowel.Dis 15 (4): 599–607. doi:10.1002/ibd.20803. PMID 19023897.
- "Voice Extensible Markup Language (VoiceXML) Version 2.1". W3C.
- "Voice Browser Call Control: CCXML Version 1.0". W3C.
- "Speech Recognition Grammar Specification Version 1.0". W3C.
- "Speech Synthesis Markup Language (SSML) Version 1.0". W3C.