The team behind VoiceGain has more than 12 years of experience in making speech recognition work in the real world. We developed speech recognition ivr applications for Fortune 500 companies in cable, telecom and high-tech and we were directly responsible for automating over 200 Million customer support calls. We built these speech applications on top of our own Dialog Engine and utilized artificial intelligence to guide the dialog, modify the call flows, and improve the recognition results from commercial speech recognizers.
When exciting new hardware developments (like GPUs) and developments in AI made Deep Neural Networks possible, we decided to work on our own DNN Speech Recognizer. Initially our goal was to augment the commercial speech recognizers that we used in our IVR. Very quickly we realized that our new customized recognizer performed better than some commercial ASRs.
This lead us to launch a full scale effort to build a Speech to Text platform from ground up. We wanted to allow developers to customize the models and make it easy for them to use speech-to-text in a wide range of applications like transcription and IVR.
We also realized from our enterprise experience that many speech applications would require deployment at the edge. This would allow clients to have full control over their data and data security. This would also minimize the total cost of ownership for large volume applications.