Hello fellow Chatbots members. I wanted to take the opportunity to let everyone know about an iPad/iPhone/iPod Touch app I have created. ChatScript 3.04 is embedded in the app, and the majority of the ChatScript coding was done by David Newyear (also a member here).

It’s on sale now at 33% off for a limited time, so this is a great opportunity to pick up a copy and check it out.

Chatty Addie is a fun app for four to twelve year-olds that listens, understands, and talks in English using voice recognition, an artificially intelligent chat engine and speech synthesis. It feels like having a FaceTime video call with a charmingly beautiful animated character.

http://chattyaddie.com

- Chris Boraski

Does it do SIRI style functions or is it just a chatbot? If so, are we able to talk to it for free before buying?

As of right now it does not do any SIRI style functions. It is strictly conversational. The vision for it is to do some SIRI-like functionality in the future, but the target market for this is kids from four to twelve years old and so the information retrieval and actual capabilities will be strongly limited to be kid-safe. Initially the plan would be to be able to retrieve information to answer some questions from a kid-safe filtered version of Wikipedia and perhaps a filtered image search as well.

As far as demoing it for free, there is no way to do that right now. I plan to put up a video soon that will show it conversing with someone so you will be able to at least see some of the interaction. The reason I can’t make it free is because I am using some of the best speech recognition and speech synthesis software and services available. This is my key differentiator, and it makes this chatbot something that is primarily a verbal interaction rather than a typed/written one, which is great for young kids that have full verbal capabilities but aren’t yet reading, writing, or typing experts.

This speech-enabling software has licensing costs, and if I made the app free, I would actually be losing money on each sale. Even at $0.99 cents, after Apple takes the 30% commission, I am hardly making any money on each app sale.

Adding advertising would be one way to allow me to make it free, but I feel strongly that advertising should be avoided in apps for kids due to inadvertent and unintended activation of ads by kids not knowing what the ad means or what it is for. Making it non-verbal would also allow me to have a free version, but that defeats the purpose and goal of this app and completely changes the experience. There is a keyboard mode that people can use, but that’s intended as an advanced option for older users and not as the primary interface.

I hope that helps. I’d be happy to answer any other questions, and I’d love to hear feedback from anyone here if you do decide to download the app and try it out.

Hey Chris,
great idea targeting an app towards that demographic. I hope you carve out a niche for yourself. I do have one suggestion. Since your demographic will more than likely not be purchasing for themselves, I would add an interactive WebUI to an instance of your app so that parents can check it out before downloading. Your actual purchasing demographic is likely to be in the borderline “tech savvy” range, or on the lower side they will be in that “too busy to buy spontaneously” category. Having a WebUI would allow them to test anything they might be nervous about putting into the hands of their kids before downloading. The video and screenshots are great (nice job by the way) but I have a feeling that parents are going to want to try specific speech patterns that they might be nervous about. From the description of your architecture it should be fairly simple to accomplish.

Best of luck

Vincent Gilbert

Thanks for the feedback and the kind words, Vincent. I could consider a WebUI, but it would have to be text-only, which might be sufficient and would at least get someone to sample what it has to offer. Maybe I would make available only limited topics for that version.

It may be better to have a free version which is text only, that way they will get a better feel for what the app does and how it works, although there is an additional step to download that even if it’s free.

I’ll weigh both options and consider them.

When you say “specific speech patterns” - I believe you are talking about expletives? There is a profanity filter which catches about 75 of the worst offenders and remaps them to innocuous similar sounding words or phrases. I don’t point that out as a feature in my marketing but maybe I should do so. If you are thinking of something else, let me know.

Thanks again!

- Chris

Rather than limiting topics or subject matter, I would recommend limiting the number of interactions or volleys (say, 50-100 per session), instead. Just something to consider.

Chris,
Expletives would be a prime example. Other people might find religious material objectionable, while others would find a lack of religious material objectionable, or references that are perceived as having some connotation that are considered objectionable to some ethnic group that other ethnic groups are unaware of, etc… (Some words in English have an entirely different meaning for a particular group than for others) Honestly it is really hard to catch what one person finds objectionable it seems. Giving them a “no pain” option for giving it a quick check, will increase sales. They used to call it “Kicking the tires”. And with reference to that, It is indeed that one step that will hinder at least some sales. No exaggeration here, Ive been building web sites since literally there were websites and there are 3 fundamental “natural laws” that I’ve come to believe are so universal that they are like web axioms.

1. People visiting your site or using your application will never, never, never, and I mean never, under any circumstances, up to and including threats of dismemberment… read the instructions that you have so painstakingly prepared.
2. People will universally feel that they have the right to complain about your applications functionality under the most incredibily [dare I say it]] moronic conditions no matter how flagrantly and egrigiously they have failed….to read the instructions .
3. The fewer steps between you and a sale the better.

Vince

Yes, this makes a lot of sense. I like the idea of limiting volleys as well. To serve this purpose, it seems a web version would be better than a free demo app because it is faster and easier than downloading. However, some portion of users might not think twice about downloading the app to try it and might be less inclined to check out the web page. So maybe both approaches are good in the long run.

I hadn’t thought much of the religious or other ethnic sensibilities but that makes sense. There is a topic about December holidays which does make mention of some religions and it’s pretty innocuous in my opinion, but who knows maybe somebody would be bothered by that.

I have to laugh about your axioms. Too true.

All of this makes me think of the line: “You can please some of the people all of the time, you can please all of the people some of the time, but you can’t please all of the people all of the time.”

About the only “axiom” that should apply here is “suck it up”. Political correctness is so out of hand right now that it’s actually infringing on personal rights and liberties. If someone sneezes and I say “God bless you”, they have no more right to complain about my choice of words than if I were to say something against… Wait! Stop! Sorry. This isn’t the place to discuss religious/political views. My bad.

Great job Chris! The app looks awesome! I was looking for an interface like “Oddcast” characters that does animated lip syncing on IOS/Android. Did you create the lip syncing characters or was it licensed from Oddcast/Sitepal? I’ve been waiting for them to come out with an IOS licensable version of their talking characters but there doesn’t seem to be any info on this. If it’s a trade secret and you can not reveal the source, I would understand as well. Thank you.

Thanks Francis!

The character has nothing to do with Oddcast/Sitepal. It was created by animating a set of custom illustrations. The text-to-speech engine (NeoSpeech) I am using supplies a real-time data feed of phonemes/visemes, which I used to synchronize the voice synthesis with the animation. It’s all custom-coded and at this time I don’t have any intentions of creating a module that could be used for this elsewhere.

Thanks Chris. The character animation and voice syncing is very nicely done. Excellent work!

Francis - There is another lip-syncing implementation I saw on the Mary TTS mailing list.  The author used it to create some videos including a talking head on a computer screen.  You may be able to get more details from him.
http://astrofra.com/wordpress/project/amiga-memories/

Thank you Oliver! I’ll check it out.