I am new to CTI and I am trying to implement CTI in my org.
currently I have CTI Server's API I can hit those API with the APEX classes.
now my concern is that how can we pass the media packets with the CTI Adaptor like if my head phones are attached with my system than I would be able to hear voice and talk as well currently I am using desktop based softphone for this, if this softphone is not registered with the CTI server then also can't manage to make a call with the given API. Now I am having two questions.
can I remove that desktop based softphone and can make Open CTI adaptor as softphone?
If I will be able to make adaptor as a softphone then how would I be able to send my Voice packet over the voice channel?
thanks advance.
Related
I search a way to control my CybovacE31 with a computer to integrate it into home
automation system. I know that it is compatible with google home and Alexa but i don't find the protocol used.
Do you have an idea how to do it?
Thanks.
My vacuum : https://kyvol.com/products/cybovac-e31
Not really an answer, but got too long for a comment, so posting it here:
TL&DR; It's likely to use a proprietary protocol between the cloud and the device. It's not public and unlikely to become public.
In case of Alexa, you need to enable a proper skill, right? It's also likely to use a vendor specific mobile app for the initial discovery that uses either Wifi or BLE. Once discovered, the device gets wifi credentials and the skill sends discovery to Alexa, so it can be controlled.
So the registration flow looks like this: Vendor Mobile -> locate device -> get wifi credentials -> send wifi credential to device -> connect device to the cloud using user account.
To control device with Alexa the Amazon account should be linked with the vender user account and skill should be enabled.
The flow for a voice command looks something like this: Echo -> (internet) -> Alexa Cloud -> Skill -> (internet) -> Device.
Why I'm explaining all above? This flow is relatively complicated and it's not enough to know the protocol, but also device cloud should trust you, otherwise you can hack the device. Therefore to control the device, you'll need to be somehow registered and trusted. For most vendors creation of such a support doesn't worth it, therefore they don't publish the protocol.
I am using ADAL on Android and I access my devices via the Graph API after enrolling them via Intune Company Portal app.
I can obtain the list of devices via
https://graph.microsoft.com/v1.0/me/registeredDevices
Is there a way to somehow identify the current device I used for making the API call and match it to one of the devices from the response?
In Intune Company Portal I can clearly see which device I am currently using. Is this possible using Graph API somehow? Thanks!
No. AFAIK this is not supported at current version of Microsoft Graph. If you require this feature, you can submit the feedback from this link.
I want to create custom Alexa skill in C# and try to get response on my laptop for running any application.
So user can ran any application by sending commands using Alexa dot.
Alexa needs the endpoints to support SSL. The usual approach here is to setup https://ngrok.com/ to tunel to your laptop and run your own application server to handle the commands. This tutorial uses RoR but the idea is the same: https://echobind.com/blog/alexa-custom-skills-with-account-linking/
I need to create a (demo) application for Google Glass with a simple user interaction: insert a card on Glass and get a response back to the application.
I think this can be done by using the Mirror API.
This application is not a web application so I think I need to use a service account.
I have created an API project on the Google APIs console https://code.google.com/apis/console/ and enabled the Google Mirror API.
After that I created a new client ID with application type "service account" (calls Google APIs on behalf of your application instead of an end-user; more info at https://developers.google.com/accounts/docs/OAuth2#serviceaccount).
The result is a client id, email address and public key fingerprint and a P12 key with password.
The problem is that I cannot find an example how to do the service account authentication and for example a card insert.
Any ideas? The used programming language is less important...
See also Can I use OAUTH2 Service Accounts with Glass Mirror API? but unfortunately without an answer.
Edit:
From the answer below I understand I cannot use the service account for this.
Is it then correct that I always need a web application where Google Glass has a callback url so data can be send from Glass to the application?
We develop a Warehouse Management System where the operator in the warehouse uses a voice client (like http://www.epf-gmbh.de/bilderorg/talkman_joe.jpg) that communicates with the server. The server sends commands to the client and the operator can send voice commands back to the server.
For demo purposes I would like to replace the voice client with Google Glass.
Edit 2:
Something like this: http://youtu.be/kbcskj4yAvo
You cannot do timeline operations with a service account. Most Mirror API operations (on the Timeline, Contacts, and Subscriptions) require a user's account since these operations must be done on behalf of the user in order to correctly identify which Glass will be used.
Update:
Your followup questions don't exactly relate to the authentication question, and they may be better asked in a new question, but two quick points:
Yes, when working with the Mirror API if you want to get information from Glass into your server you'll need a webapp which can take callbacks. You'll be registering this as part of a Subscription.
The example you pointed at uses the GDK, not the Mirror API.
I have developed a dating website built on top of the Google App Engine, to which I would like to add instant messaging, and possibly/probably audio and video conferencing.
Given that the users on the website do not want to share their personal details or real contact information, I am handling all of the login information and sessions without assuming that the clients have (or even want) a google account ID or any other login that is associated with their real identity.
I would like to hear suggestions on how I could go about adding instant messaging to my website given that I cannot just directly access Google Talk or some other existing service.
Would it make sense to use XMPP for this, and if so will Google Talk or any other XMPP service provider allow me to register new user accounts without manual intervention (ie. after a user is registered on my site, automatically register them with the XMPP provider)? Or, if not, perhaps I can use a single google ID with Google Talk with a different resource identifier for each user (me#google.com/user1, me#google.com/user2, etc...), and send messages between the different resources? Could this work, and/or would having thousands of simultaneous connections to a single account get me banned from Google Talk?
Perhaps some kind of AJAX based solution might make more sense given the fact that users are already registered on my website, but are not registered for an XMPP service?
Any suggestions about how I might approach this problem would be greatly appreciated.
Kind Regards
-Alexander
Text chat is the easier problem. You can do either with or without XMPP. Without XMPP, you'll be building a Facebook chat type client on your pages that sends messages from each user to the app, and the app then shows then on the recipient's screen.(The client can be polling, or use comet when it comes out). Check out olark to see how this works.
Once you build code to use the app as a switchboard that routes the correct message to the correct person (anonymously, maybe), you can port this easily to XMPP if you require. Both parties add you.dating.site#appspotchat.com to their buddy lists, and you send all messages from girl#site.com to guy#site.com and vice-versa. (assuming a heterogeneous site.)
Audio and video, I have no clue how to do without sharing details between the parties :-/