Over the years, dramatic changes in how we interact with computers have occurred. We went from flipping switches to stacking punch cards, typing away on hard copy and now, manipulating objects on a desktop using touch screens. Each transition has provided a greater ease of interaction. But, advances in speech recognition are completely changing the game: The idea of chatbots is decades old, but it exploded in 2016 when Facebook opened its Messenger platform to third-party developers.
Currently most of the speech-driven systems available are: 1) triggered task models (that do things for the user) or 2) call and response systems (just fun to interact with as entertainment). The trend is moving towards Intelligent Conversational UI (full-fledged dialogue systems with broad domain, a user can interact with it via speech and/or text and engage freely), that do not exist technically yet (only in movies or science fiction stories). Interfaces like Siri Alexa, Amy etc. come close to that. However, most NLI’s are domain specific and narrow, but even in that confined domain they don’t work well. Additionally, the technology is walled-off, proprietary, and secretive.
The main novelty of Pertimm’s platform in regards to existing platforms is derived from the following disruptive approach:
- Collaborative aspect of the platform;
- The semantic open-source interpretation cartridges, not only code;
- The form and genericity of the rules of interpretation: chatbots/conversational platforms currently available are based upon a two-tiers structure (intents and entities). VOQAL.AI is based upon a three-tiers structure (intents, interpretation rules, entities). This offers a power of expression that is exponentially greater than all other platforms;
- Multilingualism to progressively incorporate all EU languages;
- Automatic learning of interpretation rules by specific AI mechanisms integrated in the platform. Currently, conversational platforms help the bot developer by proposing examples of bots. Pertimm intends to take it to a whole new level by merging existing bots, recognizing that some bots can benefit from other bots in different areas (i.e. a car reservation bot can benefit from a hotel reservation bot).
- Data privacy: the IP address or mobile number of the users are never recorded.