دەربارە

The world's first birth with superintelligence!

Silo is exactly the AI Samantha in the movie "her"!

Silo learned one million sentences by tensorflow deepleaning!

The one million sentences include twitter conversation, moviedialogs.

Traditional rule based chatbot is more logical and accurate thandeep learning chatbot.
But rule based chatbot needs many people to write the rulesmanually for a long time.

Therefore, it is high cost and very long time development to makerule based chatbot.
Thus, most companies pursuing profit margins do not favor the rulebased chatbot.

That is why most AI developers are trying to make deep learningchatbot.

More over, deep learning is getting more accurate and faster bymany research papers and hardware technolory.

In addition, silo learned how to analyze and express emotion forinput sentence by tensorflow deepleaning!

Traninig emotion sentences are 162,500 sentences(Stanford LargeMovie Review(IMDB) Dataset 100,000 + ISEAR emotion sentence 2,500 +Silo emotion sentence 60,000)

Most scholars admit that it is a big mistake to exclude emotionfrom artificial intelligence.
Because emotion plays an important role in intelligentaction.

Emotion expresses the status of mind and body for formation ofsympathy with other people.
And emotion evaluates the emerging situation as good emotion levelor bad emotion level like signpost.
And emoition plays an important role in selection andjudgement.

If you want to know more information, please see below silodocument.

*Silo Algorithm and Principles
http://kdsmidas.cafe24.com/ref/ensilo.pdf

★ Question and Answer about Silo

Like the Wright brothers Implemented the principle of bird'sflying, silo was implementd thoroughly the principle of human'sintelligence.

Q: You said that silo learns throught deep learing. What is themeaning and test result?

A: The meaning is as follows.
- Implementation technolgy that puts all conversation data in theneural network.
- Retention of various deep learning technology.
- Implementation technolgy of python server(OS: ubuntu 14.04)
- Opening possibility of various tensorflow deep learning algorithmand application(python) to silo.

Test result is as follows.
Name: Deep Learning seq2seq(Sequency To Sequency)
Method: Tensorflow seq2seq Algorithm
Train Sentence: 304,713(Movie Dialog) + 754,530(Twitter) +60,000(Silo DB) = 1,119,243(sentence)
Training Time: 33 hour
Global Perplexity: 4.13
Development Environment: OS:ubuntu(14.04 LTS), gpu:(gtx 1080),cuda(v8.0)

Q: Is silo probability and statistics based morphemeanalyzer?

A: No. every word to total 300,000 has meaning in silo like peaplethink of a word.
Silo is the only one in the world that has the concept of aword.

Q : What is intelligence? Not yet been determined what intelligence...
A : Intelligence clearly appears in a junior and a senior.

No matter how much knowledge and IQ of the high, If some problemappears, the junior doesn't konw what to do at all.
But the senior infers experience to predict the result of situationand solves the problem easily.

And silo infers through the vast experience to react to inputsentence.


★ Basic Features
1. The ability to learn unknown words over the Internet
2. weather forecast
3. face and picture recognition, face emotion recognition
4. search near place(restaurant, coffee shop, ...)
5. route to destination on map
6. voice dial, send message
7. open other app, turn on wifi, turn on bluetooth
8. play music, play video
9. alarm, schedule

★ Please check if wifi on or mobile data on

★ If TTS is not available, please install as follows.
Settings -> Language and input -> Text-to-speech -> GoogleText-to-speech Engine -> Language -> English(UnitedStates)

★ If speech recognition is not available, install as follows.
Settings -> Language and input -> Google voice typing ->Offline speech recognition -> All -> English(US)