I am using Bing for speech in botframework and when i deploy it the microphone is enabled only in Chrome and Edge, I read that its supported also in some other browser?
First, the error is triggered in this piece of the WebChat code (https://github.com/Microsoft/BotFramework-WebChat/blob/master/src/SpeechModule.ts#L117)
Then, I guess that there is a combination of things.
Based on this page, the browser compatibility is:
As you can read, it seems that recognition on Firefox Desktop is not yet available.
Also, this page shows speech (recognition and synthesis) support.
Then, not sure if the following will also apply to other browser, than Chrome, but maybe yes
Related
We are developing PWA for our app because it's capable of 99% of our needs , however we do need NFC reading capabilities (not writing) and we would hate to abandon PWA and go Native just for this 1 thing
Our purpose for NFC is for reading the serial only. In fact we dont need fancy NFC functions. We just have a text field where the NFC code should be typed in, kinda like how barcode readers work (keyboard emulation)
Is this possible with PWA today?
The Web NFC API has a draft status at W3C. However, it seems to be the case that Chrome for Android has support for NFC, but only in experimental modus.
This means that currently, if you would release your application today, NFC will not generally be supported. However, the fact that Google has it available in experimental modus, seems to indicate it will be more widely available soon.
You can check the status here
For in-house solutions (or solutions where you have full control over the device) you can do it without NFC api in browsers (at least until it is available).
You will need to install Tasker app (paid, about 3 USD), which can create NFC read trigger, which then can open url like yourapp.com/{nfc_uid}. If YourApp is installed as PWA, you can select to open the url in the app instead of browser. This way it will look like the PWA is reading the NFC tag.
To make it work:
1. Install Tasker
2. Add new -> Event -> Network -> NFC Tag. Leave field blank and confirm (or just press back button)
3. Add new Task to this trigger: System -> Send intent
Action: android.intent.action.VIEW
Data: http://yourapp.com/%nfc_id
Extra: com.android.browser.application_id:yourapp
Target: Activity
4. Save
The reason we are not doing it using Task "Open url" is because we do not want to open new tab each time it is triggered.
If you have everything right and you scan NFC tag, your app should load with the NFC UID in url.
I think that you have any solution, if you don't have solution yet, so I recommender check this web site.
https://www.chromestatus.com/feature/6261030015467520#
https://googlechrome.github.io/samples/web-nfc/
to enable NFC experimental on Chrome type chrome://flags on G Chrome Android explorer
I am using following article to enable speech, which is working fine in Chrome (even with bing speech).
https://learn.microsoft.com/en-us/azure/bot-service/bot-service-channel-connect-webchat-speech?view=azure-bot-service-3.0
However same is not working in Firefox, though it has been mentioned that it should work in Firefox.
I also posted same question here, but havent got a reply, any suggestion?
https://github.com/Microsoft/BotFramework-WebChat/issues/1141
Following is the console log of firefox:
This browser does not support speech recognition botchat.js:20:29333
t
https://webchat.botframework.com/scripts/webchat-stable/botchat.js:20:29333
<anonymous>
https://myurl/speechtest/samples/speech/index.html:93:28
This browser does not support speech recognition CognitiveServices.js:15:32927
e
https://webchat.botframework.com/scripts/webchat-stable/CognitiveServices.js:15:32927
<anonymous>
https://myurl/speechtest/samples/speech/index.html:100:27
I do a test to enable speech recognition and synthesis in the Web Chat control via Bing Speech service, which can work in both Chrome and Firefox.
To troubleshoot the issue, you can try:
1) Upgrade your Firefox to the latest version, I'm using Version 62.0
2) Make sure you allow your site to use your microphone
The problem was with the js file i was using. As per thread below, I downloaded correct js file and it started working.
https://github.com/Microsoft/BotFramework-WebChat/issues/1141
I wrote a web site with google earth api and it is working fine.
I want to open this site via mobile browser, and it's giving me the following error:
"the plug-in of google earth is available only for win and mac".
How can i get through it?
Thanks.
:)
I am having a similar issue. I wish to try and use Android WebView instead of setting up GooglePlayServices at this time for an art project
I have the example loading just fine on Windows and Mac, but I need this for my Android art project.
My source code can be viewed at this link below:
www.rennakanote.com/earthdraw
I will attempt to see what I can try from Google Maps V3 APIs as per your instruction JasonM1 and see if I find a way to load this in Android.
Unfortunately, the Google Earth Plugin and Google Earth API is currently only supported on the Windows and Mac platforms.
Full list of which OS versions and browsers are supported can be found here:
https://developers.google.com/earth/documentation/index#installing_the_google_earth_plugin
The mobile version as well as the Linux version of Google Earth does not currently support the Google Earth plugin.
If you redesign your web site using Google Maps V3 APIs then it would be supported by mobile and desktop web browsers. Google Earth is a thick desktop and has restrictions especially with respect to mobile devices that a pure web application such as using Google Maps APIs can avoid.
Although the Google Earth plugin doesn't work on mobile devices, both Android and iOS devices have a Google Earth app available for them.
Thus a good way to show your data on both desktop and mobile devices is simply to publish it as a KML file, which will load in all versions of the program. The mobile devices don't actually offer a "load file" option, but instead rely on the fact that if you present the device with a KML file with the correct mime type, then the system works out that the Google Earth app is the appropriate one to display the file.
What technologies would I need to know to write an app like the now defunct Microsoft SharedView or something like TeamViewer? Any way to do it with a browser and not need a client app?
I'm a .NET developer, but figure I'd need to know C++ or driver stuff?
How would you stream the users desktop to another user? How do you even capture it in realtime?
I can imagine how you could take screenshots of the desktop and transfer them, but how do you capture live video of the screen of application and stream it to another user.
There are many apps that do this: Skype, GotoMeeting, TeamViewer, SharedView, Citrix, logmein, etc. but I'd like to write my own.
How would I get this to work on Windows, tablets, droids, etc...?
The browser seems to be a good platform for this, but there are some limitations
1 - flash doesn't work at all on IOS, and is not widely available on android.
2- Webrtc works with chrome, firefox and opera on mac/pc/linux, and with firefox/chrome on android. There's librairies to use webrtc from an IOS native app(in objective C). Screen Sharing on the other hand only works with chrome (pc/mac/linux). There's a work in progress in firefox.
3- Installation of browser plugins will be hard if not impossible on various platforms, but it can open some possibility : on chrome and firefox you can make them with javascript. For example a javascript extention can share a tab in chrome.
Using javascript you can stream from a desktop to any other desktop / android.
I'm looking into chrome frame and I'm wondering how it works.... http://scriptsrc.net shows that it's a javascript that can be embedded on the page....
http://code.google.com/chrome/chromeframe/ shows it as an install file... does the JS prompt for an installation or something?
Google chrome frame is a plugin (think Adobe Flash) for Internet Explorer, that replaces core modules of IE (unlike Flash), like layout/rendering and javascript. But the UI remains. Then there is some javascript that can detect if this is real IE or IE+Chrome, and also offer an install option to the user. Also there is a HTTP-header and a meta-tag that are required to switch IE to chrome mode, when available.
The JavaScript is to detect it and enable it if it's there. (scriptsrc is just giving you the link to the file, which is on the Google CDN.) The end user does have to do a one-time install of it, it's a plug-in (like Flash or Java). Quoting from this page in the Google documentation:
In Internet Explorer, check() determines if Chrome Frame is installed. If not, the user is prompted to install it.