An Education in E-Learning

Article Featured Image

Just because you’re part of a $4 billion business that employs tens of thousands of people across the planet doesn’t mean you need a complicated solution to create an effective e-learning environment. In Wipro’s case, all it took was a simple software add-on called Speech-Over.

Wipro, a global IT services provider based in Bangalore, India, has more than 50 development centers and 30 offices around the world. With a client base of Fortune 1000 and Global 500 companies, practice areas include business process outsourcing, technology infrastructure services, and product engineering solutions. It’s within the latter area that Wipro’s VLSI & System Design Group resides, focused on the very technical aspects of high-end chip development. With more than 2,000 team members located on multiple continents, keeping everyone up to speed about rapidly changing chip technologies and processes proved to be a costly, labor-intensive undertaking.

"Traditionally we would hold a training session in one of our developments centers, and everyone would come in from all over to that center," says Santhosh Madathil, group head of Wipro’s VLSI & System Design Group. "A domain expert would also come and deliver that training. Coordinating this meant having enough people to come out to one place, they’d have to travel, and this was a very time-consuming effort."

By August 2007, it was apparent to Madathil and his team that a Web-based learning solution would be the next step. "We were in search of an e-learning model that had no boundaries," says Madathil, who has been with Wipro for nearly a dozen years. "It needed to be available all the time, across all locations and continents."

Their search started off broadly, looking at what existing e-learning platforms were available and what kinds of capabilities were possible. One option, of course, would be to produce an entire course, replete with video and audio, in a recording studio or other venue that offered the proper equipment. The finished product, however, would be just that, with no going back. If changes had to be made after the fact—which wouldn’t be very unusual in such a fast-advancing technology sector as semiconductor design—the person who recorded the lecture would have to be tracked down again to rework the audio, which would then have to be spliced into the presentation. But that headache was better than the migraine of having to rerecord an entire presentation.

A better option? Converting text to speech (TTS) and then embedding it into an actual presentation. "First we learned to see what was possible, then we looked for something to handle that," Madathil says. "In doing that search, we came across Speech-Over through its Web site."

Speech-Over is a PowerPoint add-on from Israeli firm Tuval Software Industries that integrates Microsoft’s popular presentation software with TTS. "We were quite used to creating our own training materials and presentations in PowerPoint," Madathil says. "Speech-Over used the same framework."

Speech-Over, which came to market in 2005, operates in PowerPoint XP, 2003 or 2007. In the XP version, it appears in the PowerPoint menu bar. Equipped with TTS capability, Speech-Over accepts any voice that’s compatible with Version 5 of the Speech Application Programming Interface (SAPI). Using Tuval’s Voice Effect Marshalling technology, Speech-Over lets course developers automatically synchronize slide narration with screen animations, allowing them to describe and explain text and graphic objects on a slide in any order to create a multimedia training movie.

"Speech-Over short-cuts the business of recording narration in e-learning courses. It reduces something that is time-consuming and complicated to text-based authorship," explains Joel Harband, who founded Tuval Software in 1990, though only in recent years turned his attention to speech applications. "As you’re working within Power-Point, the speech commands are right there inside."

Here’s how it works: In PowerPoint, you select a picture, chart, bullet point, etc., and click on the Add or Edit button. Speech-Over’s Add Narration Clip dialogue box pops up, where you type in the text that should accompany the selected screen item. Next, you choose a voice that’ll narrate the text, click on OK, and the spoken text is embedded within PowerPoint at that point within the presentation. During playback, the voice is synchronized with the graphic.

Developers also have the option of recording their own voices and importing prerecorded sound files. The entire PowerPoint presentation is then converted into Flash so that it can live on the Web, Harband says.


Like any smart customer, Madathil gave Speech-Over a try before making a decision to buy. "We did an evaluation and were actually quite happy with its features and ease of use," he says. Madathil’s original purchase was for five licenses, costing $164 apiece; at the time, Tuval was running a special so that Madathil received another 10 licenses for free. Madathil opted to buy AT&T Natural Voices TTS software along with Speech-Over "for a more professional-quality voice." For PowerPoint-to-Flash conversion, he chose FlashSpring, another PowerPoint add-in, from Russia-based CPS Labs.

Ironically, creating the actual material for the presentations took longer than tracking down and purchasing an elearning software solution. "We were first creating the courses from scratch," Madathil says. "That took two-and-a-half months. But the tool was ready to go immediately."

While using the tool, Madathil and his team came up with a few ideas for improving Speech-Over—suggestions that Tuval integrated into the software and put in subsequent product releases that Wipro received. Those features included pause and stop functions, an autosave function, and a refresh function.

"I rely on my users to point out issues," says Harband, who is planning to offer Speech-Over in a centralized server version (to date, Speech-Over is sold on a per-license basis and includes a year of free maintennce and updates).

The VLSI & System Design Group’s first e-learning course was completed and posted to its intranet in December. To date, 12 presentations have been developed and 30 Speech-Over licenses have been purchased. Since making revisions is considered standard operating procedure, Madathil especially appreciates how simple it is to change the voice-enabled learning material. "Traditionally you have to go back to the source and edit it," he says. "Here making the revisions is very easy. You just go in, write or edit the text you want to change, and create a new version."

In addition, the savings in time and effort needed to coordinate and administer training have been considerable, according to Madathil. "The cost [for Speech-Over] is like a one-time investment," he says. "But having a learning platform available to all of our team members that can be easily accessed anytime from a desktop is, for us, more important than any cost savings involved."

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues