I am looking into a way to integrate limited-vocabulary speech recognition into my Android app (since Google Voice Search relies on Internet access which isn't always available to users).
I found a Nuance page enticing programmers to download NDEV or the "Dragon Mobile SDK" but if it requires Internet access just like Google Voice Search, why should I be even interested in it?
Unfortunately, Nuance's page is not very informative and requires you to register (and sign a dragonian agreement...) before you get a chance to spend time to download & play with the kit only to discover that it's not what you're looking for...
Do you happen to know whether the speech-recognition part of NDEV or he "Dragon Mobile SDK" require Internet access?
Do you know of any off-the-shelf SDK for Android that doesn't require Internet access (i.e. some limited-vocabulary speech recognition can be done locally on the Android device)?
Yes, NDEV does require Internet access. Their SDK is just an API wrapper + the UI with the microphone button (and the code necessary to capture the audio properly).
If you need local speech recognition, have a look at Sphinx, it's a great project.
On most phones (with hardware powerful enough) Google's own speech recognition can be set to work offline.
Simply go to Settings->Language and Input-> Voice Search -> Offline Speech Recognition
There you'll be able to download data for any language you need available offline.
Related
I want to build an app to transfer message history between iOS and Android.
Apple does not allow third party apps to access messages, but Samsung Smart Switch circumvents this by using either phone-to-phone USB connection or by accessing iCloud. Google phone transfer seems to do a similar thing.
How do these apps make this possible and is this something third party developers can do?
I don't think you can transfer things over like that from Android to Apple iPhone unless they give you the option when you are first setting up the iPhone to transfer contents over from android and it also transfers the messages.
Another option though it's been a long time since I have used an Android device. You could use something like https://messages.google.com/web to load up your android messages from your phone and possibly just export them and save them to your computer.
As for developing such a tool, working within the messages API for apple. I think you can only develop extensions or addons but can't directly access the content for security reasons. Without further research, the companies might be only allowing the company to do this via a contract between them, but would not let a 3rd party vendor such as you or I access this data.
https://developer.apple.com/app-store/review/guidelines/
Nowhere in the guidelines does it specify that you can't access the SMSes. But you can only access it if you use private methods which are not allowed and will get your app rejected.
You can only access data for which Apple supplies a documented API. You can not access files outside of the Sandbox of your App unless Apple provides an API for it.
Can I build custom actions to leverage the Google Assistant SDK without needing to be online? I'd like to have the natural language parser to help me navigate my Android App, even when offline. That is, speak to it and say special words to navigate the screens and what not, using the voice instructions of the user.
This seems totally possible using the Assistant SDK on the internet, but I need the app to run on my android even outside of connectivity.
Google assistant uses google servers so I don't think it is possible, but there are some other options check out this link.
Android: Speech Recognition without using google server
My question is regarding custom voice commands for Android.
Is there a way to develop an app, not necessarily a UI app, that adds custom voice commands to any Android mobile?
For example, after installing AccuWeather on Android, every time I ask Google Now for the weather, it displays information fetched from the said app.
Did AccuWeather added new voice commands to my phone?
Can any app developed by anyone add custom voice commands to an
Android phone?
Can I write and install an app that, triggered by the voice command
"How is my portfolio?", fetches data from the stock market, returns it
and then be displayed in the same view, the Google Now view?
I Googled all of these concerns but found more questions than answers.
You can write an app that registers itself as a choice for the available system commands.
As of this writing, you can not register an app that registers a custom command. Google used to accept requests to register custom voice actions, but stopped. I don't know why they stopped, or if/when they'll allow customization again.
BTW there was a possibility with an XPOSED Module to add custom commands to google search ( and so you can use the voice recognition even with a smart watch connected ), but it stopped working since they updated the google search app.
if you are just doing this for personal use u can still use it with an old version of google search app and this xposed framework module and correct hooks ( they postet the correct hooks for some updated versions of google app later in the thread , just use the latest one...).
here you go:
https://forum.xda-developers.com/xposed/modules/mod-google-search-api-t2554173
i really hope google will bring out a better voice API than the actual is.
Google has recently made great progress with their speech recognition software, which is used in several open source products, e.g. Chromium Web Speech and Android Handsfree texting. I would like to use their speech recognition as part of my server stack, however I can't find much about it.
Is the text recognition software available as a library or package? Or alternatively, can I call chromium from another program to transcribe some audio file to text?
The Web Speech API's are designed only to be used in the context of either Chrome or Android. There is a lot of work that goes on in the client so there is no public server to server API that would just take an audio file and process it.
If you search github you find tools such as https://gist.github.com/alotaiba/1730160 but I am pretty certain that this method of access is 100% not supported, endorsed or confirmed to keep working.
The method previously stated at https://gist.github.com/alotaiba/1730160 does work for me. I use it on a daily basis in my home automation programs. I use a python script to capture audio and determine what is useful audio or just noise, then it sends the little audio snippet to google and returns the text all under a second!! I have successfully integrated it into my programs and if you google around you will find even more people that have as well!
Google has speech recognition services available for use from mobile phones (Android has it built in, iPhone users can use the Google application) - http://www.google.com/mobile/. We've found one article where someone tried to reverse engineer the service at http://waxy.org/2008/11/deconstructing_google_mobiles_voice_search_on_the_iphone/.
We want to better understand what is happening over the network when we use Android's RecognizerIntent. Does anyone have any experience using this service over the web or know of other articles that may explain its workings?
I read this presentation few weeks ago- http://www.abelski.com/courses/android/speechinput.pdf
The following link is a 3 mile high review of the Google Voice Server ....
http://www.google.co.jp/events/developerday/2010/tokyo/pdf/tt1-gruenstein.pdf
Answer: just move your .apk file to your android phone it will work. Error is only occuring because we are trying to do in emulator