 |
|
|
 |
|
SpeechFirst is probably the most sophisticated speech interface available on the market. SpeechFirst uses Microsoft SAPI 5.1 TTS and ASR engine for voice recognition and text output, SpeechFirst incorporates many complex layers of parsing data, enabling it for command recognition, categorizing it, and providing the user a pleasant user interface. SpeechFirst also provides the ability for users to create their own grammar (dictionary) so as to recognize commands not normally recognized by speech engines. |
|
|
| The Given |
|
| About 30-man years of R&D done by a contractor had yielded a product that was ill-designed, bulky and not workable outside the laboratory environment. At the time the product was known as SpeechBand. The niftiest feature in SpeechBand was that it was a browser plug-in and created a separate URL bar for surfing the net in speech mode. The underlying technology design did not reflect the finesse or the functionality conceptualized by the customer. |
|
| The Needs |
|
| The need was to create a Voice Users' Interface (VUI) that could voice enable a web page on the fly. While most any text to speech reader could read out the textual content, none of the ones existing at the time could take voice commands. Another differentiator desired was that the page should be categorized on-the-fly so as to create a mental map of the page being accessed. |
|
| Building the software |
|
| After months of trying to patch up the SpeechBand, our engineers tossed out the entire design and went back to the drawing board. It took some brilliant engineering talent and fresh, out-of-the-box thinking that resulted in the SpeechFirst we see today. " With a billion or so possibilities, the complexity involved in creating a technology such as this is mind-boggling " said a CTO that we had invited to review SpeechFirst. |
|
| We learnt that the web wasn't a perfect world and that web design practices were such that a standard approach could not be taken. We did compromise on the dream of cent percent on-the-fly speech enabling, but we have a technology platform that continues to evolve. |
|
| While building SpeechFirst, we conceived SpeechBench, a Voice authoring tool that could voice enable web content for telephony access. |
|
| SpeechFirst is designed using C, C++ and Visual Basic. |
|
|
|
|
|
