patrickjquinn / p-brain.ai Goto Github PK

Heads up: https://github.com/Picovoice

The Cheetah subsystem would be most relevant here.

Some users may want to have this application externally accessible. The app could probably use SSL support and authentication, although that may be tricky over socket.io.

Get errors after yarn & yarn start:

c:\HomeAssistant>yarn start
yarn run v1.15.2
warning package.json: License should be a valid SPDX license expression
$ node server.js
Using default SQLite database provider
Setting up database.
Error: Cannot find module 'c:\HomeAssistant\node_modules\sqlite3\lib\binding\node-v72-win32-x64\node_sqlite3.node'
Require stack:

• c:\HomeAssistant\node_modules\sqlite3\lib\sqlite3.js
• c:\HomeAssistant\db\sqlite_db\index.js
• c:\HomeAssistant\db\index.js
• c:\HomeAssistant\server.js
  at Function.Module._resolveFilename (internal/modules/cjs/loader.js:610:15)
  at Function.Module._load (internal/modules/cjs/loader.js:526:27)
  at Module.require (internal/modules/cjs/loader.js:666:19)
  at require (internal/modules/cjs/helpers.js:16:16)
  at Object. (c:\HomeAssistant\node_modules\sqlite3\lib\sqlite3.js:4:15)
  at Module._compile (internal/modules/cjs/loader.js:759:30)
  at Object.Module._extensions..js (internal/modules/cjs/loader.js:770:10)
  at Module.load (internal/modules/cjs/loader.js:628:32)
  at Function.Module._load (internal/modules/cjs/loader.js:555:12)
  at Module.require (internal/modules/cjs/loader.js:666:19) {
  code: 'MODULE_NOT_FOUND',
  requireStack: [
  'c:\HomeAssistant\node_modules\sqlite3\lib\sqlite3.js',
  'c:\HomeAssistant\db\sqlite_db\index.js',
  'c:\HomeAssistant\db\index.js',
  'c:\HomeAssistant\server.js'
  ]
  }
  (node:13976) UnhandledPromiseRejectionWarning: Error: Cannot find module 'c:\HomeAssistant\node_modules\sqlite3\lib\binding\node-v72-win32-x64\node_sqlite3.node'
  Require stack:
• c:\HomeAssistant\node_modules\sqlite3\lib\sqlite3.js
• c:\HomeAssistant\db\sqlite_db\index.js
• c:\HomeAssistant\db\index.js
• c:\HomeAssistant\server.js
  at Function.Module._resolveFilename (internal/modules/cjs/loader.js:610:15)
  at Function.Module._load (internal/modules/cjs/loader.js:526:27)
  at Module.require (internal/modules/cjs/loader.js:666:19)
  at require (internal/modules/cjs/helpers.js:16:16)
  at Object. (c:\HomeAssistant\node_modules\sqlite3\lib\sqlite3.js:4:15)
  at Module._compile (internal/modules/cjs/loader.js:759:30)
  at Object.Module._extensions..js (internal/modules/cjs/loader.js:770:10)
  at Module.load (internal/modules/cjs/loader.js:628:32)
  at Function.Module._load (internal/modules/cjs/loader.js:555:12)
  at Module.require (internal/modules/cjs/loader.js:666:19)
  (node:13976) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). (rejection id: 2)
  (node:13976) [DEP0018] DeprecationWarning: Unhandled promise rejections are deprecated. In the future, promise rejections that are not handled will terminate the Node.js process with a non-zero exit code.
  Done in 3.54s.

This library is so cleanly written! I didn't once consider picking up my ax.

You took a largely confusing subject and very cleanly described it in code so even I (one with no NLP experience) can read and understand it.

I'm confused on one point though, in api/ask-core.js

if (confidence > 0.25) {
    throw new Error('error')
}

I know that the confidence value there is a value from 0-1 describing how sure the algorithm is in classifying the query as the returned label, but why does confidence need to be less than a quarter? In my lay-mind it looks like a statement, "If you are more than a quarter certain of this result, error out."

Thanks for any clarification you can give!

I was planning to make a connection using sonoff smart wifi switch. So my question is that how to connect it with the IoT skill assuming that the switch should locally connected to wifi

Hi Guys,

So i've been giving some thought to the best way to return 'rich content' such as weather cards, news etc and have come up with a partial solution that I would like your feedback on.

Basically, we generate a html card with the weather forecast say on the server (for example in the weather skill), then render that to a canvas and finally a png file and rend that along with the response. That way we can have rich responses that can rendered on any client, be that web or native, by simply rendering the image.

Does anyone have any feedback on this?

Hello to anyone that is listening. I have installed P-Brain.ai with no issues on Kubuntu linux and when i speak nothing happens, it say's that the web browser is using my mic but nothing happens. i've tried also in firefox etc. I've also tested my mic with arecord so it works. What should i do or is this a glitch with Kubuntu 18.4.
Best Regards and hope to hear from someone.
also can email me at uboat478@gmail

Respectfully Bogs

I would like to know how to change the hotword from brain to something else in your codebase and also please redirect me where I have to place the snoyboy pmdl file. I am not able to see that file in the repository but in the video tutorial, it's there. Please do needful help.

NOTE: I have tried modified in src/index.html but it doesn't work for me.
Thanks for the great tutorial.

Hi,

I removed all skills except the time and it started throwing the exception below. Any thoughts?
Thanks

λ yarn start
yarn run v1.5.1
warning package.json: License should be a valid SPDX license expression
$ node server.js
Using default SQLite database provider
Setting up database.
Setting default global values in database
Loading skills.
Training recognizer.
TypeError: Cannot read property 'value' of undefined
at classify (C:\temp\P-Brain.ai\api\core-ask.js:123:34)
at classify.next ()
at onFulfilled (C:\temp\P-Brain.ai\node_modules\co\index.js:65:19)
at C:\temp\P-Brain.ai\node_modules\co\index.js:54:5
at Promise ()
at co (C:\temp\P-Brain.ai\node_modules\co\index.js:50:10)
at toPromise (C:\temp\P-Brain.ai\node_modules\co\index.js:118:63)
at next (C:\temp\P-Brain.ai\node_modules\co\index.js:99:29)
at onFulfilled (C:\temp\P-Brain.ai\node_modules\co\index.js:69:7)
at C:\temp\P-Brain.ai\node_modules\co\index.js:54:5
(node:31144) UnhandledPromiseRejectionWarning: Unhandled promise rejection (rejection id: 2): TypeError: Cannot read property 'value' of undefined
(node:31144) [DEP0018] DeprecationWarning: Unhandled promise rejections are deprecated. In the future, promise rejections that are not handled will terminate the Node.js process with a non-zero exit code.
Done in 2.22s.

The roadmap on the Wiki needs to be fleshed out with @patrickjquinn's goals for the project to guide contributions.

How do i tell brain to open youtube or gmail etc. Also to play movies just returns a black box that doesnt seem to open any movie. Is my OS missing something?
Best Regards
Bogs

After following steps (of installing yarn), I still get follow error when I run yarn start Error: connect ECONNREFUSED 127.0.0.1:4567. Do I need to specify the API keys before I can use a simple command (like hello Brian)?

I am running P-Brain.ai in Docker

When verbally asking Brain for the time Chrome shows
resp_types.js:91 aborted

When using HTTP it responds normally
{"msg":"It is currently 5° and Clear","type":"weather","question":"{what is the weather}"}

I apologize I am not able to provide clearer logs, but given some instruction I would happily provide them. Thanks for your awesome project!

P-Brain not working with nodejs v12.14.0, it will be an update soon? so we can continue working the project with latest features... many thanks!

`C:\WorkSpace\brain>yarn start
yarn run v1.16.0
warning package.json: License should be a valid SPDX license expression
$ node server.js
Using default SQLite database provider
Setting up database.
Error: Cannot find module 'C:\WorkSpace\brain\node_modules\sqlite3\lib\binding\node-v72-win32-x64\node_sqlite3.node'
Require stack:

• C:\WorkSpace\brain\node_modules\sqlite3\lib\sqlite3.js
• C:\WorkSpace\brain\db\sqlite_db\index.js
• C:\WorkSpace\brain\db\index.js
• C:\WorkSpace\brain\server.js
  at Function.Module._resolveFilename (internal/modules/cjs/loader.js:797:15)
  at Function.Module._load (internal/modules/cjs/loader.js:690:27)
  at Module.require (internal/modules/cjs/loader.js:852:19)
  at require (internal/modules/cjs/helpers.js:74:18)
  at Object. (C:\WorkSpace\brain\node_modules\sqlite3\lib\sqlite3.js:4:15)
  at Module._compile (internal/modules/cjs/loader.js:959:30)
  at Object.Module._extensions..js (internal/modules/cjs/loader.js:995:10)
  at Module.load (internal/modules/cjs/loader.js:815:32)
  at Function.Module._load (internal/modules/cjs/loader.js:727:14)
  at Module.require (internal/modules/cjs/loader.js:852:19) {
  code: 'MODULE_NOT_FOUND',
  requireStack: [
  'C:\WorkSpace\brain\node_modules\sqlite3\lib\sqlite3.js',
  'C:\WorkSpace\brain\db\sqlite_db\index.js',
  'C:\WorkSpace\brain\db\index.js',
  'C:\WorkSpace\brain\server.js'
  ]
  }
  (node:92036) UnhandledPromiseRejectionWarning: Error: Cannot find module 'C:\WorkSpace\brain\node_modules\sqlite3\lib\binding\node-v72-win32-x64\node_sqlite3.node'
  Require stack:
• C:\WorkSpace\brain\node_modules\sqlite3\lib\sqlite3.js
• C:\WorkSpace\brain\db\sqlite_db\index.js
• C:\WorkSpace\brain\db\index.js
• C:\WorkSpace\brain\server.js
  at Function.Module._resolveFilename (internal/modules/cjs/loader.js:797:15)
  at Function.Module._load (internal/modules/cjs/loader.js:690:27)
  at Module.require (internal/modules/cjs/loader.js:852:19)
  at require (internal/modules/cjs/helpers.js:74:18)
  at Object. (C:\WorkSpace\brain\node_modules\sqlite3\lib\sqlite3.js:4:15)
  at Module._compile (internal/modules/cjs/loader.js:959:30)
  at Object.Module._extensions..js (internal/modules/cjs/loader.js:995:10)
  at Module.load (internal/modules/cjs/loader.js:815:32)
  at Function.Module._load (internal/modules/cjs/loader.js:727:14)
  at Module.require (internal/modules/cjs/loader.js:852:19)
  (node:92036) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). (rejection id: 2)
  (node:92036) [DEP0018] DeprecationWarning: Unhandled promise rejections are deprecated. In the future, promise rejections that are not handled will terminate the Node.js process with a non-zero exit code.
  Done in 1.00s.`

Hello,

I want to play with this, but I can't while at work unless it has a license.

What license do you intend for this code?

Awful Javascript error is awful.

{ responseType: 'fact',
  originalQuery: 'heavy rain move youtube' }
TypeError: Cannot read property 'length' of undefined
    at get_wlfra_result (/Users/anonrig/Documents/forks/P-Brain.ai/skills/fact/index.js:36:27)
    at next (native)
    at onFulfilled (/Users/anonrig/Documents/forks/P-Brain.ai/node_modules/co/index.js:65:19)
    at process._tickCallback (internal/process/next_tick.js:103:7)

If /api/ask is used without tokens, then a new token is created each time it is called. This isn't fatal but could get quite messy.

At the bottom it reads "Ester eggs", I think you meant "Easter eggs". ;)

Hey there, I just wanted to make sure you were aware of the offline speech recognition library sonus which uses Snowboy for offline key-word detection.

I personally run ubuntu and had to do a bit of googling to make sure everything installed properly but overall it works rather well for what I use it for. :)

Hello @timstableford @patrickjquinn Please help me to understand
Could you help me and tell me how to enable or view server logs, collect and run the docker container like in your readme. my knowledge is lacking, I’ll be thankful to you

Hello, I am integrating and because to log in it is necessary to do it from a pop-up window, in electron it is not possible to do it. I would like to know what is the fraction of code that is responsible for displaying this pop-up window and how I could do to login with HTML.

I'm getting the message "unauthorized".

Multiple return is not possible so is there a way to get multiple responses
For Example. Me: Download movie
Brain: Download has started.
on download finish it should say. Brain: Download Completed

I try to add Russian, he recognizes everything well, but when I add a keyword in Russian to the weather module, he doesn’t see it and all the time says he doesn’t understand. You can not tell what needs to be corrected so that everything is recognized, maybe some kind of regular expression needs to be changed, or somewhere else corrected from en-UK to ru-RU

Hi Guys, i'm currently planning a re-structuring to make this project more accessible and add a strictly defined structure that will foster further growth without adding layers of confusion & complexity. This is the first step toward making P-Brain a truly intelligent system.

The proposal is to re-structure the project to mirror that of the human brain. It could potentially look as follows:

Current Project Structure:

/api/
src/
src/
log/
docs/
skills/
test/
config/
server
yarn.lock
tremux.sh
README.md
Dockerfile
package.json

Proposed Project Structure:

lobes/
	/parietal - maps request to skill
	/temporal - Stores any memory based task such as name, location, own name etc
	/occipital - always running, self training which is constantly running and learning from responses and requests
	/frontal - natural language interpretation, acts on commands and queries 

stem - Contains the interface of the server, i/o for request:response pairs

cerebellum - any management task such as thread management, timing or synchronized activities live here

neurons - skills, extendible etc


yarn.lock - Package Management 

scripts/
	  /tremux.sh - Running the project on an Android device end-to-end

README.md - Project info
Dockerfile - Docker config 
package.json - Package and script manifest

I just need a confirmation if windows version working or its broken. I Sync my 'old' version to the newest one just to realise it not working (though it starts a server but it gets stuck on the 'wait a min' bubble) and npm/yarn giving me errors

I have mysql provider, and the index.js in /db/mysql/ dont insert or get the data in my mysql database. How can i enable mysql?

i got a "Failed at the [email protected] install script." error when running "npm install" i got this error and do not kn ow how to fix got any suggestions

Hi guys, I want to include offline speech recognition that's open source to this project, initially just for English. It might be worth investigating the work of XNOR.ai and failing that, building a full, optimised model for PocketSphinx.

Longer term id like this model to to be trained by interactions with the platform and then have some sort of central repository for the model so it can be synced across all instances of the platform.

Any one willing to help? Or have any ideas?

Cool project!

Got a few questions. Will you set up gitter or something for chats about the project?

Does the system always-listen or does it deduce a hotword locally? Since it would be nice if ti weren't listening always ("snowboy" software could be used for local hotword generation).

It uses Google for STT and TTS, right?

For video 4 you could maybe display a skill that communicates with a device, for example MQTT protocol is the most popular protocol for home automation for makers, for controlling devices...So would be nice if you showed a MQTT skill.

Could it work "headless"?

Also how come it doesn't detect the "hotword" or the sentence in the first go in your video? You have to say twice all the time. Maybe you could buy a PS3 cam (it is very cheap) which has a 4-mic array, that one will give you way better audio.

What about interacting with the linux-shell or starting programs (mplayer with an url pointing to a mp3 podcast for example)?

Why i say hey Brain - nothing working? Mic enabled

npm install in termux script not working now, perhaps a problem with the nodejs version.
How did the script work for you, on which version of nodejs? In the description, you have 6, but in termux there is only 12

Hello,

I would like to know if i can help to support multilanguage (french for me) and how can i do.

Thanks.

I'm on windows 7, I installed node js and yarn and started the project. The server is up but when I say a sentence, brain replies with the same sentence. Here is the error I got. Is there a way I can share more debug information. I'm noob with js.
Awesome project by the way, cheers!

D:\gitlab\P-Brain.ai-master>yarn start
yarn start v0.19.1
$ node server.js
P-Brain is listening on port 4567
{ Error: ENOENT: no such file or directory, scandir 'D:\gitlab\P-Brain.ai-master\api\skills'
at Error (native)
at Object.fs.readdirSync (fs.js:951:18)
at getDirectories (D:\gitlab\P-Brain.ai-master\api\core-ask.js:15:13)
at train_recognizer (D:\gitlab\P-Brain.ai-master\api\core-ask.js:26:14)
at next (native)
at onFulfilled (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:65:19)
at D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:54:5
at co (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:50:10)
at toPromise (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:118:63)
at next (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:99:29)
errno: -4058,
code: 'ENOENT',
syscall: 'scandir',
path: 'D:\gitlab\P-Brain.ai-master\api\skills' }
(node:4892) UnhandledPromiseRejectionWarning: Unhandled promise rejection (rejection id: 2): Error: ENOENT: no such file or directory, scandir 'D:\gitlab\P-Brai
n.ai-master\api\skills'
192.168.0.72
TypeError: Cannot read property 'value' of undefined
at Object._query [as query] (D:\gitlab\P-Brain.ai-master\api\core-ask.js:48:26)
at next (native)
at onFulfilled (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:65:19)
at D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:54:5
at co (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:50:10)
at toPromise (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:118:63)
at next (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:99:29)
at onFulfilled (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:69:7)
at D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:54:5
at co (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:50:10)
TypeError: Cannot read property 'value' of undefined
at Object._query [as query] (D:\gitlab\P-Brain.ai-master\api\core-ask.js:48:26)
at next (native)
at onFulfilled (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:65:19)
at D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:54:5
at co (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:50:10)
at toPromise (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:118:63)
at next (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:99:29)
at onFulfilled (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:69:7)
at D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:54:5
at co (D:\gitlab\P-Brain.ai-master\node_modules\co\index.js:50:10)

Just creating this to track the next short term task. Currently sessions/tokens are registered just to a user, if they were registered to a device it would allow some interesting commands such as "On my laptop play Everybody Knows".

Changes necessary:

• Add device list table to database, where the primary key is a combination of the device unique ID and user ID. Devices well also have an optional nick name. Device unique IDs will be IMEI numbers, browser fingerprints etc.
• When each token is created it's registered to a device in the list. If there's already a token for that device and user, the old token is deleted.
• Add two commands 'Call this device qqqq' and 'On my qqqq do qqqq'.

Hi im looking to get into ai and coding and when I saw this I wanted in so I installed everything (on windows) and went to run it but when I did yarn start it just referenced me here: "https://yarnpkg.com/en/docs/cli/run" can you help?

Hey folks,

So I want your feedback on what new skills, and skill types you want to see in the default P-Brain skills. Some ideas include, IoT protocol based skills, Gmail, Calendar, Slack, IFFT and Social Media.

If you have any notions, ideas or concepts on how what subsystem work needs to be added in order to build more engaging skills, let me know.

I've got a question about using skills. I'm trying to write a food/restaurant skill using the EatStreet API. How do I add a continuous chat? Example:

You: Find a restaurant.
Bot: Okay, where are you?
You: "location".
Bot: Great! Here are some restaurants near you! ...

Right now, once the bot has returned text, it's not listening to something else said by the user. Is there a way I can get this to work? Thanks!

The Geolocation API used by the weather skill now seems to be a premium service, which has broken the skill.

i have js code

const natural = require('natural')
const speakeasy = require('speakeasy-nlp')
const genify = require('thunkify-wrap').genify
const fs = require('fs')

const classifier = new natural.BayesClassifier()
natural.BayesClassifier.load = genify(natural.BayesClassifier.load)

I need all this to work with Russian letters. Help with this please.
Example of the Russian language if required: Привет

//set PT language; natural.Language('PT-BR');

but when I use it my docker doesn’t start up maybe I’m doing something wrong.Help me.

Hi Pat , I changed the hot word Brain to Harry in src/index.html file as you instructed in the comment section of your first video but nothing happened ,,Can you please help

Hi all,

Myself and @patrickjquinn plan to move all the separate configs to a central database, that we'll provide the same API for no matter the result of this.

Behind the scenes, what database system would you like to see as the default database? Would you like something simplistic like dbdata which stores in JSON? Or maybe something more complex like sqlite3 or MySQL?

Tim

Hello @timstableford @patrickjquinn Please help me to understand
Could you help me and tell me how to enable or view server logs, collect and run the docker container like in your readme. my knowledge is lacking, I’ll be thankful to you

Hello, This is the text of my json processing module.The problem is: when I specify the port, it stops working and the assistant says that he did not understand me.
How to fix it???
my knowledge of the code is not strong, you maybe need to connect another one library.
but I do not understand how to connect them, I work through docker.
help me please

constrequest = require('co-request')
const util = require('util')
const intent = () => ({
keywords: ['number qqq', 'number action qqq'],
module: 'abonent_oblgas'
})
function * abonent_oblgas_resp(query) {
const request = require('request');
let country_res = request("http://192.168.10.55:8444/getabonentbynumberofcontract?number=10&id_podr=1")
country_res = JSON.parse(country_res.body)
const loc_string = country_res.city + ',' + country_res.country
return loc_string.toLowerCase()
return{text: 'works.'}
}
const examples = () => (
['number 587', 'number action 144']
)
module.exports = {
get: abonent_oblgas_resp,
intent,
examples
}

How can I upload and make online on hosting using amazon AWS.

Hi, how can i remove the login? i want do it with HTML.

how to run the project on the latest version of node js, what needs to be rewritten or is it impossible ????

I get the following error on osx (node v7.7.1, Python 3.5.2)

$ yarn & yarn start
[1] 31664
yarn install v0.21.3
yarn start v0.21.3
$ node server.js
warning [email protected]: License should be a valid SPDX license expression
[1/4] 🔍 Resolving packages...
success Already up-to-date.
✨ Done in 0.66s.
Setting up database.
TypeError: Cannot read property 'version' of undefined
at getVersion (/Users/suhajdab/Dropbox (onereason)/projects/_useful/P-Brain.ai/sqlite_db/index.js:84:15)
...

Any ideas how to solve it?

Help rewrite the fact / google module under Russian letters, it doesn't work, what's the problem?)

changed only index.js in the fact folder, did not change anything in the fact / google folder
Maybe need to change some kind of regular expression ???

For the more vague skills like song, they'd be better suited to hard-coded rules rather than classification. It should even improve accuracy for modules that aren't hard-coded.