Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 9 years ago.
Improve this question
I'm making an android app in Eclipse and i want to record my voice, which will be password for login into application. When i try to log in it should recognize my password and let me use the application. How can i do that comparison in order to get a match, i need something like shazam? Thanks for any tips!
Audio comparisons is a very complex topic. Generally, if you don't know anything about it, I'd discourage you to head into such a project.
The problem is, that while you could of course just compare the two audio files byte by byte, that certainly is not what you want. Although two audio files sound the same, i.e. it contains the same spoken words, the actual data will differ quite a bit.
You'd have the following possibilities:
Try to recognize what the user said (Speech recognition), and check whether the same word was recognized later. This solution, while being the simplest, could not distinguish between different users.
Dive into the mysterious world of audio processing. A technology called Fast Fourier Transformation is more specifically what you'll be ending up dealing with.
Related
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 3 years ago.
Improve this question
I have an output file with a C1B extension from a smart card reader for truck drivers DDD.
The goal is to read this file and present it as a readable data for users. The thing is that there are several apps live that can do this but so far after couple days of research I still got nothing to work.
similar app: https://play.google.com/store/apps/details?id=com.azsoftware.tachomobile&hl=en_US
Any help would be appreciated, thanks.
There isn't an implementation on android for parsing DDD's, your only solution is to implement the parser yourself, witch is a pretty challenging task. I happen to work on a project regarding tachocards myself, we have a library on elixir and it took several months to be developed.
Basic parsing by itself is not complicated, the complicated part is to parse the records array in a performant way.
A good starting point for you is: https://github.com/L1L1/cardpeek
Take a look at the source code where tachocard is parsed, there you can find very much information, like offsets and number of bits you need to read.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 3 years ago.
Improve this question
I'm about to propose a graduation project in Deep Learning for Arabic Natural Language Processing field. However, since I'm a beginner student in this field, I have already started learning a helpful series of courses, and I have almost a full year to work on this individual project. So, what are some great ideas for an Android application in DL for Arabic NLP?
P.S. I will gather the needed dataset if it's not available.
Edit: Recently, I read about various problems in this field, but most of them are very classic choices, and it seems to me that they don't give a good impression of my efforts while there exist many well-known and very efficient similar applications such as Google translate and Siri. However, Since my question is still too broad, kindly write down your suggested edits and recommendations.
NLP is a broad area and have lots of application. And you will also need the supporting applications(web pages, editor, reviews in arabic). I will give some ideas and try order them easy to difficult.
Spell check
Sentiment analysis App - read reviews(movie/restaurant etc ) and rate them(provided you have data (sentence / +/- ))
FAQ from websites - given a question extract answers from FAQ of the website, we can NER to get organisation name and question similarity measure.
Grammar correction - require lots of data(wrong - correct sentences) and editor app to use it in android.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 7 years ago.
Improve this question
I'm working on an Android app (though eventually I'll want to do the same thing on iOS) and I'm looking to build an image recognition feature into it. The user would snap a picture, then this component of the app would need to figure out what that image is, whether it's a bowling ball, a salad, a book, you name it. It would also be helpful if it could figure out roughly how big the object in question is, though I imagine the camera focus values could help with that. The objects in question would not be moving.
I've heard of neural networks being used, but I'm not sure how this could be implemented, especially since I want to be able to recognize a very wide range of objects. I highly doubt this sort of processing could happen natively on a phone either. What are some solutions to this problem?
I would suggest you look at OpenCV. They have an awesome open source library for image processing and object detection. They also have great Android sample apps ready for testing some of their APIs.
http://opencv.org/platforms/android.html
Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 8 years ago.
Improve this question
i be wondering How to game calculate the time for completion of work?
e.x:clash of clans find something update are done.
i believe someone want say get the time by server.but even i be offline can receive notification which say done.even i do change clock of phone nothing change in game if i change time of phone.
The time that is used to determine the yield from objects existing in clash of clans is maintained from a centralized server.
If you wanted to build an app that would calculate the times required to do things in game I think the rout to go would be to
A) ask the developers for the code (unlikely to be fruitful)
B) reverse compile the code. (potentially illegal)
C) Hand document the times required for in game events to happen (game w/notes)
D) Do a literature review to find data on the game
From this Data you may be able to craft a sort of clash of clans advisory app, but I doubt it will be easy to dupe such a consistently maintained app with hacks as simple as changing you're clock.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 8 years ago.
Improve this question
Is there any API in Android SDK which should recognize an object and return the name of the object.
For general objects in 3D, that is an unsolved problem in computer vision right now. A lot of researchers are working on it, but right now computers cannot reliably identify that an arbitrary object they haven't seen before is a "chair", for example. (If you think about it, such labeling actually requires a lot of judgment and world knowledge to know what kinds of things humans can sit on, and that's beyond the current state of AI for objects in general.)
There are algorithms that basically do a Google Image Search: they take a given picture and use some fairly advanced computer vision to find similar-looking pictures on the web (i.e. Google Goggles). There are APIs for those; check out
Google goggles API
Those work well for 2D pictures, like posters and product logos, that always look exactly the same, but not for things like plants and animals.