I keep getting direct questions which broadly translates into one single question, “What IVR tool/Software should I use for my IVR requirement ?”

Over period of last three years, I can classify, three types of people ask me such questions, they are :-

1. Representatives of companies who is looking for IVR systems and they also have an IT department.
2. Fresh Engineers who want to form a start-up company in voice applications
3. Students looking for projects while completing their course

While I always try to use encourage students to go for using APIs available in C++ with the CTI hardware, it is really difficult to suggest about IVR tools or IVR solution for type #1 and #2 above!

As for students, I feel it would do them good if they learn about those C++ APIs in order to understand the telecom protocols too!

I find it difficult to suggest an IVR solution for the type #1. Here are my few points #1 should consider while choosing an IVR solution:-

1. Do they have at least one qualified telecom engineer as well as one software developer available full time to develop IVR solutions, if not, on should seriously consider hosted IVR solutions with pay-per-usage option. It will really help them remain happy if they can get the right IVR hosting provider. Also this would be cheapest option without having to worry about CTI boards, Telecom lines etc.

2. If they have manpower available, choosing the right tool may not be difficult! Most of the proprietary IVR tools like VBVoice, OsiDial etc. normally support Dialogic CTI Hardware, while open source Asterisk, FreeSwitch support low cost CTI hardware like Sangoma, Digium. ExtendIVR is one such tool which supports almost all CTI hardware available in the market!

But even if one can decide on the IVR tools to be used, the worry of investment in CTI board and telecom connection will still be there!

3. If the IVR requirement is very specific which requires lot of integration with existing customised software, one does not have much choice then going for a custom developed IVR solutions from an IVR company. It might be costly affair though.

So, according to me, these three are the thumb rule to decide on a IVR solution for type #1 above.

I think, it is most difficult to give any suggestion for the type #2. I still do not know what to suggest for them. Still, I think here are few guidelines:-

1. VXML is getting standardized and getting popular now a days. So any tool which ca support VXML may help building an IVR solutions which will be more popular in coming days. But the competing IVR tools which uses traditional development tools like C#, VB have advantages of readily available manpower and minimum training period.

2. If one is considering providing hosted IVR service, then consider about investments in CTI hardware as well as its support. If this is a constraint or plays a major rule in decision making, one will have to choose an IVR tool suitable for such CTI boards only. For example, if one goes for low cost Sangoma boards, it is best to go with Asterisk in Linux.

3. It is normally assumed that the budding entrepreneur would be a good programmer too, but this is not not necessary! Still, I would say the third guideline to decide on would be programming skill as well as knowledge on telecom domain.

This is my personal observation and I would welcome any suggestion from experts in IVR.

One of the most irritating aspect of IVR ( Interactive Voice Response) System is having to listen to the complete prompt before making a choice. While this is essential for first time dialer to the IVR system, it is really irritating and time consuming for regular callers who already know the options. So, many IVR systems give an option to press option Keys during the announcement is going on. While using speech enabled IVR where commands are entered through voice, the same procedure is followed.

So a good IVR System should be intelligent enough to recognise the first time caller and frequent caller by capturing and storing Caller ID. This may be useful specially in customer support IVR provided in Telecom companies. It may increase usability of IVR to some extent. Also, there are many IVR applications where entering data by DTMF is much faster, accurate and desirable.

But this same “Key Press Through” can be a reason for other type of problems in some cases. In many practical IVR implementation, multiple Keys (DTMF – Dual Tone Multiple Frequency) or digits required to be captured in the form of PIN ( Personal Identification Number ), Credit Card Number, Customer ID number, Mobile Number, Date format in DD-MM-YY format, amount etc.

These multiple digits entries are normally terminated by pressing HASH key (#) or Star (*) Key or some time just maximum allowed digits or maximum allowed inter digit pause.

For example, it is always a practice in IVR to get the multiple numbers or values entered by IVR user confirmed by pressing 1, to renter press 2. So in a scenario of entering Customer ID of 5 digits length, IVR caller is allowed to press 12345# or just 12345. In both ways, if the IVR user presses the numbers correctly and wait for some response from IVR, it is good, but what happens when a user presses like 123451 continuously or 12345#1 continuously ? Chances are, if the press through is enabled, the IVR will skip the next confirmation step having received the key 1 already pressed and user will not even hear the confirmation prompt!

Also, it is important to decide how the error prompts a well as time out prompt would be played depending on the number of digits entered. Many clients may want to play an “Invalid Customer ID” prompts when a caller enters 123451 as it exceeds maximum length of Customer ID. Explaining these small but important features, options to customer might become cumbersome when clients do not understand them!

But it will also serve as shortcut for the smart IVR user who can bypass the irritating confirmation step by already pressing 1 along with the Customer ID.

So, in case the IVR wants the user to listen out the confirmation prompt, IVR developer can disable “Key Press Through” feature while playing Confirmation prompt, while allowing it in longer menu option prompts.

This is a small details while developing your IVR applications, but this can improve usability of the IVR and also improve accuracy of data entered by user. It will also make smart and regular IVR users happy as it provide them a shortcut to avail the information provided by the IVR.

What is predictive dialer and what is Click To Call ?

Predictive dialer is such smart automatic telephone dialer system which dials out a telephone number from a list of telephone numbers in order to reduce holding time for both agent in call center as well as the person who is called from it. Though there are various techniques to implement predictive dialer, the basic aims remains same :-
1. Called party should not kept on hold;
2. Agent in call center should not waste time in hearing RBT ( Ring Back Tone) of the called number.

Click To Call and Predictive Dialer

Click To Call Service is a one to one conference where the called party ( as in above Predictive Dialer Definition) initiates the call and speaks to human agent immediately! So the immediate purpose of reducing wait period for both called party and agent is automatically achieved!

Now, does that mean that Click To Call service would (should) replace predictive dialer in order to maximize use of finite and costly resource like outbound telephone call charges as well as time of human agent ?

The first consideration is predictive dialer is not used just as a call back service where some user or web visitor will register their phone number (submit their phone number by filling up a form) expecting a call back from the website or some call center of the company who owns the web site.

Also, predictive dialer can dial from any phone list, not necessarily gathered from web site visitors, but also from off-line registrations, manual phone number collection, collecting phone numbers by door to door marketing etc., not to mention collecting of phone numbers when a user provides his phone number when registers for service, like a DTH ( Direct to Home) TV service, registering for credit card, bank account etc.

While click to call service essentially collects numbers from website visitors only. Apart from traditional Click To Call service, now a days, smart implementation of click to call where a call is imitated as soon as a web form is submitted website visitor, which acts as an additional feature of immediate call back service on an existing registration form! ( Niche Tech Service private Limited provides such customised Click To Call service. )

So, most probably, it will not be possible that Click To Call Service would replace predictive dialer!

Both Predictive Dialer and Click To call service can have options to choose the time when a person expects the call back from the website or call center.

Predictive dialer is also used to make scheduled outgoing calls which is not either standard function in a Click To Call service or available only at extra cost or when a Click To call service with IVR ( Interactive Voice Response) functionality! But now a days, many Click to Call service providers can provide complete IVR functionality or solution in a Click To Call service too.

So, a click to call system, along with smart and customizable IVR at the back end of it, can be surely a next generation of Predictive Dialer and can potentially dethrone it from future call center!

Another important point to consider is, predictive dialer can no longer be used to blast out calls to random numbers after strict implementation of “National Do No Call registry” ( NDNC) in order to stop unsolicited calls. Click To Call service facility eliminates the requirement NDNC registry before making calls. But then, spammers can surely misuse Click To Call!

I had posted about possibility of replacing predictive dialer with outbound IVR, and now I feel, and and IVR system with outbound feature and Click To Call functionality can surely replace aging predictive dialer and bring in cost saving as well as ‘Cloud Telephony’ in Call Center world!

Presently, outbound IVR ( Interactive Voice response) System is widely used in India as well as around the world by candidates in any democratic election in order to reach their electorate in the form of bulk voice call or broadcasting voice messages. In India, many political parties deploy Bulk Voice service during election campaign asking people to vote for their candidates.

But instead of using IVR only to broadcast voice message, political parties can also use inbound IVR in various ways to connect with their electorate regularly, instead of reaching out only during election. Now a days, even in India, many political parties are using websites to register members online, collect donations along with informing achievements their party agenda, achievements, leaders etc.

How inbound IVR can be used ?

Political parties can use inbound IVR for few specific ways. Some of them are :-

1. Register New members with phone numbers

Inbound IVR can be used to register new members with verified phone number. The phone number detected by Caller ID can be used as a verified phone number by the new registrant which is better than online web based registration where fraudulent registration may require manual verification. For more about IVR registration process, one may click the link.

2. Providing Information about the political party

Caller to the IVR number of the political party may be provided with options to listen to election manifesto, vision documents, party constitution as well as latest news about the party. It can be IVR portal to provide any kind of party information in voice format to any one having a telephone or mobile connection.

Apart from normal news, IVR could also be used to provide clarification to any controversial news about the party so that their is no confusion on the controversial topic in the electorate. This will increase confidence in the party as well as remove any malicious rumour.

3. Listen to speeches by key party leaders

Some important speeches delivered by the party leaders, be it in the parliament/senate or in an electoral meeting, could be provided in the inbound IVR for listening by party supporters as well as any citizen. This will help connect to more and more people.

Recorded voice by party leaders could also be accessed by party workers for specific information from time to time.

4. Complaint Registration and Complaint Status.

Another important use of Inbound IVR is to record complaints, grievances by the electorate against party candidate or any other. This will make the party more and more transparent, democratic. Complaints can also be treated as feedback to the political party. Elected representatives from constituencies may have dedicated IVR number or menu for lodging any grievances. This can make party more popular among voters.

5. Conducting Polls/Surveys

Inbound IVR can also be used to conduct surveys/polls about any policy among its supporters. This kind of IVR survey or poll can also be used to allow electorate to choose their candidate for particular constituency. Mechanism like one Caller ID being allowed to vote only once could be implemented in order to prevent fraudulent voting.

This kind of anonymous voting/survey encourage grass root level party workers as well as general citizens to voice their opinions which may help party to choose right policy, right candidates!

Apart from listed above, I think inbound IVR could be used many ways by Political parties. So, most probably, it is time Political parties start using IVE beyond bulk voice messaging.

Many a times, I get request to help installing SPCI4/SPCI2 board installing in Windows 2003 Server. While installing mine, I prefer to ask Dialogic vendor to help remotely install and configure ( normally they provide free Remote Installation and Configuration, and they always do good job) the board, but many a times I also try configure myself. So, I thought I would try to list a step by step instruction for installing and configuring Dialogic Septel boards.

Dialogic SPCI4/SPCI2 Installation

While for beginners, all these terms like SPC, DPC, SS7, MTP License, ISUP License etc. may sound too technical and scaring, but these are just parameters which are needed to be filled up in two simple configuration files.

So here it goes :-

1. Try to get a new machine and install Windows 2003 Server fresh. Also install latest service pack. Please ensure you have at least two PCI slots available.

2. Download the latest version of DPKWIN.EXE from Dialogic Website. If you do not have the link, get it from your Dalogic Vendor.

3. Install the Septel Card in any of the available PCI slot.

4. Start the machine and let windows boot. If Windows Plug And Play detects the SPCI4 board, do not press Next, press cancel.

5. Also download Dialogic® PowerMedia™ Host Media Processing Software Release 3.0WIN from Dialogic Website. Demo license is available for free which works 45 days. After that you have to download new license again.

6. Now install the DPKWIN.EXE. It will install itself in C:\SEPTEL directory. After successful installation, go to windows Command Prompt and then to C:\SEPTEL. Type net start septel and press ENTER. You should get Service started message or “Service Name not found”. In that case, the DPKWIN.EXE version does not match with your SPCI4 board. You may try downloading older version of DPKWIN.EXE.

You can stop the Septel service by using the command : stop service septel.

6. Once you get “Service Started successful” then, you may think that the board is detected successfully and running now! The next exercise will be for configuration of the board.

You have two files to edit,

1. system.txt
2. config.txt

As the name suggests, system.txt is used to instruct which modules, protocols to be executed by the SPCI board.

config.txt contains the parameter required to communicate with remote telecom switch. Few parameters which of importance and must in config.txt files are

1. DPC (Destination Point Code)
2. OPC (Originating Point Code)
3. SIF (Signaling Information Field)
4. SLC (Signaling Link Selection)
5. Timeslot with link
6. CIC values.

After the configuration is done, one has to start the SS7 link by running gtload -d command on command prompt. After it loads successfully, open another command prompt, and run the command mtpsl act 0 0 .

These values are provided by the telecom service provider. Most of the time, telecom service provider uses their own jargon for the above values which may create confusion.

The protocols like MTP3, ISUP, MAP have to be bought from Dialogic separately and they provide license button which needs to be installed in the board. Without these licenses, the system stops after every 10 hours.

Depending on application requirement, both the system.txt and config.txt has to be modified accordingly.

Most probably, the debate on merits and demerits of using ASR ( Automatic Speech Recognition) would continue for some time to come, but in case of India, there are three compelling reason avoid using ASR where simple, old, robust and universally understood ‘keypress’ can do. While ASR can surely speed up and increase interactivity of an IVR system, in Indian scenario, it can be a big nuisance, specially for IVR vendors providing IVR to be used nationwide.

Here are my reasons:-

1. Poor Recognition due to background noise

With due respect to all ASR engines, India is a loud country with lot of noises all around! So, with lot of background noise, it is not possible to recognise words or phrases with high degree of accuracy. Indian cities are noisy with lot of noise pollution all around. Even Indian houses are not well ventilated. Indian villages are crowed with various domestic animal who create lot of sounds.

Because of this background noises, good ASR engines also fail to perform properly and end up with poor accuracy. But users will never appreciate the ‘genuine’ technical problems behind this, but the blame will fall squarely on IVR developer or IVR Vendor.

2. Languages and Accent

India is home to 1652 languages. Out of these, 29 languages are spoken by more than more than 1 million people and another 122 are spoken by more than 10,000. It has officially recognised 22 languages. English is widely spoken and understood all around the country but India also has more than 250 million illiterate people.
So with this statistics, one can easily figure out the changes in accents for speaking even English numerals, forget about English words or Phrases!

This lack of common accent for speaking out any word or phrase prevent any ASR engine detecting them accurately. But, everyone who has dialed into an IVR system, already know how to press a key from the telephone instrument or the mobile. So, in order to wider use of IVR as well as accuracy, ASR should be avoided.

3. Poor Mobile Signal

Most of the recent telephone users in India are mobile phone users. Apart from big cities, large areas as well as large pockets of cities too, have poor network signal and keeps fluctuating rapidly. Due to which, two human may be able to somehow communicate, but in case of IVR, there chances that it would miss some voices, part of voice which are vital for recognising a word or phrase by ASR.

So, even the best ASR would fail to detect words accurately resulting in caller dissatisfaction, call abandonment which will defeat the very reason for using an IVR system.

While there would be remedies, tricks and techniques to address these problems, but these are top reasons if some one wants to avoid using ASR in India!

IVR system is not just a clubbing together of state of the art CTI boards, high end servers, very easy to drag and drop tool design IVR call flow for flexibility. One of the saddest part about IVR development is, the users may not appreciate about these technological break-through like SS7, ISUP, CPA ( Call Progress Analysis), accuracy and speed of speech recognition or the or intricacies involved in detecting disconnect tone, engage tone.

They simply care about what menus they hear and if the options they press or speak works! They are least bothered about the hard work some one may have put in to detect her key presses, or recognise heavily accented English numerals, but only if the IVR retrieves the information it promised in its menu. The appreciate only when they can understand the information read out to her listening first time without having to press a key to listen again!

It may not be possible to satisfy all the users of an IVR system, but by using well scripted menus in simple language and intelligently constructed sentences may surely help majority of the people to use IVR system without having to learn the IVR before using it.

The answer to all these problems faced by IVR users and making IVR more useful and user friendly is ‘scripting of the IVR menus’! Script writing for IVR means, writing the IVR menu options in simplest language, well formed sentence so that the listener understands the message accurately without having to concentrate.

IVR scripting can be used in three areas to improve IVR usability:-

1. Scripting the menus
2. Dividing complex menus to simple sub menus
3. Scripting the phrases used for providing information

1. Scripting the Menus

This is quite straight forward. The menu options where IVR gives some options to the IVR user to either press telephone keys, or speak some word or phrase, should be properly scripted. Accuracy of the options is of course a must and forgone conclusion. For example, if the menu says, “To transfer your call to a customer care executive, please press 9″, then on pressing 9 by the caller, the IVR should transfer the call to customer care executive, not to the main menu!

2. Dividing complex menus

Script writers will divide complex options or large options will divide into some groups, and make it two level menus, in place of one. This will not only help first time callers, but also make it more user friendly faster with using ‘cut through’ key pressing!

3. Scripting phrases

Now a days, IVR has to retrieve many information from various locations through either database look up or by using API. Also, it may have retrieve information in bits and pieces from different servers, then knit them together, and then finally produce a sentence which would present a complete required message/information to the IVR user! Now that may not be easy for an IVR engineer! So, one must use a professional script writer for writing these phrases either to be recorded or used with TTS.

Without proper scripts used in IVR menus and in information, IVR will not be usable at all, even by expert IVR users! It will be very frustrating for users and all the great technologies used to design and develop IVR system will completely go waste!

There are few parameters or features of IVR which are important and are common to almost any type of IVR application. These parameters play very important role in making the voice application more user friendly. These parameters are used at many stages in IVR software or in any call flow. I hope, knowing these would help people who are looking to develop an IVR or looking to purchase IVR software.

These are very small parameters, but they may play crucial role in reducing call abandon rate in IVR. Some of them are :-

1. Allowing or not allowing caller to press a key while IVR plays the menu options

IVR may be configured to accept or to reject any key press while caller is listening the menu options. Allowing to press key during playing of menu allows caller to ‘cut through‘ the IVR menus and reach his desired IVR stage or menu quickly. But it may also increase chances of wrong key press, specially when some menus options are changed. Many regular caller may not like to hear known menus again and again.

2. Duration to allow a caller to press any key or voice command

Waiting time for callers key press or voice command may become crucial. During wait period, IVR does not play anything and become silent, so if this wait period is too long, caller may think the IVR is not responding, so disconnect the call.

3. How many times to play a same menu and number of attempts to choose correct menu options

Naturally, people make mistakes while pressing key or may make mistakes intentionally. So, it is important to decide the number of attempts one may try for each and every menu options. This kind of parameter may be required even while accepting some valid input from caller, for example mobile number, customer ID, complaint ticket etc. If this is not defined, then some caller may occupy some channel of the IVR unnecessarily.

4. Confirming menu menu selection of digit (s) pressed or voice command spoken

It is an important decision to make. Should IVR confirm every menu choice made by caller or not! Even in any software, when some one tries to do some important task, it pops up an confirmation window, similarly, IVR also ask confirmation for menu choice! But it may also irritate IVR caller. When using ASR ( Automatic Speech Recognition), confirming every option may become a necessity rather than a choice.

5. Inter-digit pause while expecting more than one DTMF keys as input from caller.

While caller enters multiple digits as input to the IVR, for example, credit card number, phone number, ID number, the duration of pause in between two consecutive DTMF digits or key press is important. Very long inter digit pause may result caller disconnect.

6. Using # ( hash) or star (*) keys to terminate

Deciding to use # or * keys to end entry of multiple digit is a good option. But these keys may also confuse some people who do not know about # or * keys as many of them do not use these keys. In India, many land line phones do not have # key. But with penetration of mobile phones, one may expect people to know about these keys.

These parameters are quite common and many developers use some configuration files or provide user friendly forms to configure.

Recently, I have been getting many messages from this IVR blog visitors asking how to choose right IVR system for their organisation. While majority of the people who asked wanted to automate their customer service process, they have one common requirement, minimum initial investment possible!

IVR (Interactive Voice Response) Systems are available in two ‘forms’, hosted IVR and On Premise IVR. Choosing right IVR solution among these two types, may not be easy. I would try to document few points to consider between the two IVR types.

1. Cost of IVR system
Every IVR system has three cost components apart from general maintenance and up-gradation cost.

a. Cost of CTI hardware as well as Computer Systems
Normally this is fixed and one time cost unless some capacity enhancement is required in future. There are many options for CTI boards now a days, low cost and high cost. For example, Synway, Sangoma at the low cost end and Dialogic at the high cost end.

b. Cost of Customised Voice Application
This is the IVR call flow implementation cost which may include integration with third party application like CRM, Web Services etc. Now a days, many DIY ( Do It Yourself) tools are available in visual or simple TAG based format. For example, visual tools from Invox, Omnivox3d from Apexvoice, Osidial from Ositron, Comsys etc. and VoiceXML, PPHVoice. But one has to invest in either in-house qualified manpower or get the voice application developed from IVR software provider.

c. Telephone Cost
While incoming call is free any telephone number, still there is rental cost associated with even normal telephone connection. Toll free numbers, outbound IVR call are charged ‘per call, per minute’ basis to the IVR owner. Many IVR involves call transfer and it may also cost per call, per minute basis.

In an hosted IVR system, cost component 1.a does not arise but it can not escape the cost components 1.b and 1.c.

In on premise IVR system, 1.a would be a must along with 1.b and 1.c, but it can escape from call transfer cost by using PABX system. In a call center scenario or where human agent has to interact with IVR caller, the cost of ‘call transfer’ could be quite high in the long run depending on call volume!

2. Flexibility of the Voice Application or IVR Call flow

This is a important factor for choosing an IVR system. There are some IVR application which have fixed call flow and there are some where call flow needs to changed or configured continuously. Flight announcement, Train status, IVR payment are few of IVR applications where the call flow can not be changed being public utility. There are IVR surveys, which need to changed or configured after completion of any survey!

Even IVR in call center need not be changed frequently as it may confuse regular caller. But call center IVR needs many ‘call transfer’ for human interaction with IVR caller.

So, one has to choose between hosted IVR or on premise IVR depending on voice applications running on the IVR system. It may influence the cost of the IVR as well as its maintenance and flexibility.

So, according to me, these two factors should mainly decide what kind of IVR sysetm one should opt, either hosted IVR or on-premise IVR.

Call transfer from interactive voice response system to human agent is one of the basic requirement and a must for IVR usability in any call center. While transferring an active call on an analog extension is very simple an straightforward, but the same may not be simple in case of ISDN PRI connectivity with PBX. The method of call transfer vary from one PBX manufacturer to another and that makes it more difficult to implement call transfer facility for a new CTI application developer.

Recently, I had faced similar difficulty while trying to implement call transfer facility on ISDN PRI with a PBX. It took many days, but later I realised that, it is mainly due to technical support from PBX manufacturer end only. In my opinion, PBX manufacturer should provide information how call transfer facility has to be used by a CTI application in their manual itself.

But I have learned a lesson and I thought I would share my experience how call transfer should be implemented using D300 Dialogic board or DN300 Donjin board with a PBX on ISDN PRI connection.

I would suggest these steps:-

1. Discuss with PBX technical support clearly that you are attempting to implement call transfer function on the ISDN PRI link, not on PABX extension line. Call transferring on PBX extension line is quite straight forward, put the phone on ON HOOK for 500 ms ( FLASH time, it may again vary from one PBX manufacturer to another), OFF Hook again, dial the number where call has to be forwarded, then disconnect (put of ON HOOK ) the call. I have faced this strange reply from PBX support as soon as I asked them about call transfer facility in their exchange. If they provide the correct information, then your life has become easy, if not then you would go through the same ordeals which I had gone through.

2. If PBX support provides you with standard call transfer method like SSCT( Single Step Call Transfer) or ECT ( Explicit Call transfer) etc. which are standard and known method, then you can let out a sigh of relief, if not keep your finger crossed! They may simply inform to send some messages on on D channel sequentially and then drop the call! Also, they may send you exact message format! Now you are on your own. But you are near to the solution.

3. If you have not learned how to capture ISDN trace for CTI boards, the time has come. Dialogis has very useful and powerful ISDN tool called ISDIAG.EXE ( normally found in Dialogic\Bin directory) which can be used to capture all messages on D channel and use PRITRACE.EXE in the same channel to decode the messages in simple, human readable English language. Also, capture every error messages while using gc_SndMsg() as well all events from PBX end. If by simply sending message using the Global Call protocol function gcSndMsg(), you are able to transfer call then you have been very lucky, if not continue reading next.

4. Once you have all these trace and logs, you would sit down and study them. I am sure, you would now understand what messages you are sending to PBX and why PBX is rejecting your message or not understanding! Major problem occurs sending Facility Message or Information message! At least, in my case, I was sending a Facility Message to PBX while asking PBX to hold the call, while it was expecting a simple Information message! Normally, you could not expect PBX support guys to give this information, hit and trial is the only method.

5. If you complete the above four steps, you would be able to use any features provided by PBX which is accessible on ISDN PRI! Finally, here is the piece of working code snippet for call transfer on ISDN PRI using a Dialogic D300 CTI board and Coral IRIS PABX (using Global Call Protocol and gc_SndMsg function):-

a. Send Hold call
IE_BLK info;

info.length = 3; // Length of data
info.data[0] = 0x2c; // Keyboard Facility code
info.data[1] = 0×01; // Length of facility
info.data[2] = 0×52; // ASCI R

GC_IE_BLK gc_info;
gc_info.gclib = NULL;
gc_info.cclib = &info;

rc = gc_SndMsg( ldev, crn, SndMsg_Information, &gc_info );

b. Send Number to be dialed Information
ie_blk.length = 6; // length of data
ie_blk.data[0] = 0×70; // called party number code
ie_blk.data[1] = 0×04; // IE length
ie_blk.data[2] = 0×81; // numbering plan number type
ie_blk.data[3] = 0×34; // digit “4″
ie_blk.data[4] = 0×32; // digit “2″
ie_blk.data[5] = 0×35; // digit “5″ 425 is th number where call would be transferred

rc = gc_SndMsg( ldev, crn, SndMsg_Information, &gc_info );

c. Drop the call normally
gc_DropCall(crn,GC_NORMAL_CLEARING,EV_ASYNC);

That is it! This would transfer your call to another PBX extension 425.

I thought I would share it here as I spend considerable time trying to implement call transfer with that particular exchange and on ISDN PRI interface. I got many help from various forums ( specially Dialogic Developer Network) and Dialogic support.