Written by: Jeh Daruvala, Founder and CEO of Yactraq Online
Internet data is concurrently accelerating across many channels, with audio, video and text based channels growing faster than ever. From a business intelligence perspective, this means companies need to be listening across all channels like Phone Calls, Tweets, YouTube, Facebook, Broadcast Media and Web Pages. Applications of the business intelligence stemming from these channels include Voice-of-the-Customer, Media Monitoring, Competitive Intelligence, and Brand Sentiment, yet integrating all these channels is a significant challenge. All channels (Text, Audio, Video) are vital to understanding the overall information signal generated by consumers, and it is evident that legacy methods of tuning into this signal cannot keep pace with the rapid evolution of these channels.
All traditional approaches like interviews, agent notes, call audits, etc., that try to improve business effectiveness are inefficient, limited, and often expensive; as a result, much of the signal is lost.
Automation based on stand-alone Natural Language Understanding (NLU) methodologies is limited to text channels, and completely misses the rich data contained in audio channels. Integrating a text based NLU system with an independently built audio or video analysis system on a case-by-case, customer-by-customer basis is clearly not the right design approach.
Automated speech recognition (ASR) represents a potential, but limited, solution, and legacy systems don’t deliver competitive advantage or cost efficiencies. Legacy phonetic speech systems contain very small vocabularies and customizing these vocabularies is both expensive and time consuming. As a result such products limit themselves to spotting keywords, and not being speech-to-text systems, such products are incapable of semantic analysis through natural language understanding (NLU).
New solutions are needed to scale data analytics across traditional and new digital channels.
Yactraq’s Comprehensive Intelligence approach is to indiscriminately index all relevant data, be it text, audio, or video. This involves automatically meta-tagging every minute of audio and video from any number of channels or media platforms, and once indexed, this data is now machine searchable across all channels. As a result we can automate the consolidation of omni-channel reporting into a single dashboard, thereby revealing deeper signals and delivering actionable, high impact business intelligence. Accurate B2B speech systems need to understand product/brand names, as well as industry and company specific business terms. This requires the creation of mass customized speech vocabularies that are unique to each customer.
In the past such vocabularies were manually constructed, but at Yactraq we focus on automating the creation of custom speech vocabularies through various artificial intelligence techniques.
Yactraq’s patent pending and award winning semantic platform fuses both speech recognition and text analytics into a single core. Through our proprietary technology we have built custom vocabularies consisting of tens of millions of sentences, but can also detect very specific terms as well. All this while helping customers save millions of dollars on building such custom vocabularies, as well as opening up entire new markets that have been undeserved by legacy solutions due to prohibitive costs.
Check out our comprehensive business intelligence capability at our website, whereby we analyze data from multiple channel types (Phone Calls, Tweets, YouTube, Facebook, Websites, Chat Logs) with a single pattern recognition engine, using unified taxonomies and linguistic data.
Yactraq is an IPFD partner of Intellectual Ventures. This places Yactraq in a unique position in terms of IP strategy and a freedom to operate vs. other players in the speech marketplace. We are also among Econtentmag.com’s Trendsetting Products of 2015, have won many other accolades like TiE50 and CIX Top 20, and are an alumnus of the Canadian Technology Accelerator at Rocketspace.