Overview of InTalxys Product Range (dialogue)
Intelligent Dialogue Systems Solution Range
- Text only dialogue -chat style dialog
- Speech only dialogue -talk and listen
- Spoken dialogue with 3D avatar – talking with a visible person
- Spoken dialogue with 3D avatar and user tracking – talking with a visible person who also sees you
- Multimodal dialogue – all options available depending on the situation
Talking with a quite intelligent computer system, no longer science fiction, today reality. Use it for your business, for the sake of
- efficiency and
- client satisfaction
Don’t lag behind the comptetition, be ahead. You have the options to
- just chat in text
- converse by voice and/or
- have a dialogue with a 3D avatar which looks very human and natural, you listen and look to his/her face
- he/she can also look at you and react on your eyes, face and body language
Natlanco can develop and customize just the virtual personality you’d like for your needs, with the right kind of
- intelligence, friendliness and patience
- expert knowledge in the domains you want
- good looking, funny and addictive personality
- speaking and understanding the languages you want
Text only dialogue – the common base
The most basic dialogue option, and the common basis for the more sophisticated options
The option with only textual communication is very much like chatting with a real person.
This possibility doesn’t limit the intelligence of the virtual personality,
- while maybe the best for certain environments for the device of your clients, for instance just exchanging SMSes using an inexpensive mobile phone, getting the widest possible audience,
- ideal for a noisy environment
- or for situations requiring the highest achievable understanding accuracy. Can indeed include some automatical spelling or grammar correction. (suboption)
It also might be the easiest entry level to start exploring Natlanco’s dialogue systems
Several suboptions determine which NLP/NLU Pandula sophistication layers are active while understanding the user.
The actual dialogue system and derivatived options are internet server based, so in a way the Intalxys is SaaS (software as a service), except maybe for the client side.
Speech only based Dialogue
This option is telephone-conversation style, so without visual communication. You talk, the virtual person talks back, you understand each other.
This might be the optimal option for when a screen of some kind is not available, or not handy to use, for instance when you’re walking without VR glasses, or indeed, when just making an old fashioned smartphoneless voice call into the system.
Natlanco’s technology allows to interface to the best available speech recognition technology available for any language and environment, so that opens up a world of effectively working voice communication.
And of course, the intelligence of the system doesn’t suffer too much from not visually communicating.
Speech Dialogue with speaking 3D realistic avatar
This option allows talking to a realistic virtual person, who you see speaking back, which gives the conversation an awesome ressemblance to a ‘normal’ conversation.
There are many possibilities for the looks of the virtual person and some extra options to increase the natural impression of this smoothly animated character, like eyeblinking, nodding, context-adaptive mood of the face, breathing and lively body moves, as appropriate. As the animation normally is generated on a cloud server, the local hardware of the user is not so critical for generating high fidelity natural animation in sync with the voice sound.
Spoken Dialogue including Virtual Personality and Visual User Behaviour Tracking
This is the most ambitious option: It allows bidirectional communication in the auditive and the visual field. In other words, voice and eye/face/body language communication with the 3D avatar(s). This might be more useful than you think at first glance.
In a natural converstion it’s important to read each other’s mood, attention, focus of attention and of course presence and absence.
This option requires at least one high resolution camera on the user’s device. There are many sub-options to choose from. We are still developing/integrating several components of this option using both our own and externally, the best available technology.
Even if the most ambitious option for dialogue communication is chosen, it’s likely that not all of these options are optimal for every situation.
Instead it’s better to offer the choice or automatically adapt the dialogue communication options, suiting best to the user environment and device.
This is what is meant by multimodal dialogue.
So the same user, with a single subscription or license to the service, can:
- call in by phone,
- sms with the system,
- chat using an internet cafee screen,
- converse by speech (talking+listen)
- using a simple tablet
- or up to a sophisticated 3D installation at home,
and have in all these situations a similar conversation with the system, from the content perspective.
In modes that include a rich graphical displaying, it’s obvious that related graphical objects like lists, tables and pictures can be shown as part of the full picture with animated avatars.