Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 4 years ago.
Improve this question
I'm working on a polyphonic music transcription project. I have read some papers and gone through articles which explains similar tasks. I'am very confused about many aspects of the problem domain. Hope someone will be able to help me.
So far I have obtained a stream of decoded audio data from a given mp3.
I have understood onset detection is the first step towards transcription. . Is there any java library available which can be used for detecting onsets.
Next, detecting the fundamental frequency is also done with the use of FFT as I have read.
I want to know the use of FFT in these tasks. (I'm not familiar with FFT much). Is it absolutely necessary to perform FFT for onset detection and F0 detection.
If not what are the other ways.
Can I perform the FFT on the audiostream I have, or is there some other operation that has to be done in order to manipulate these audio data.
Thanks a lot.
This field is known as machine listening.
Polyphonic transcription of digitally encoded music is one of the holy grails of machine listening. It is an unsolved problem, and an area of active research. The sub-fields include:
Onset detection
Beat extraction (detection of the metric structure, time sig, etc)
Pitch detection (possible using auto-correllation, and other methods, on monophonic signals, but an unsolved problem when applied to complex polyphonic music)
Key detection (key signature detection).
Depending on the nature of your project, you might find it useful to explore the SuperCollider programming environment. SC is a language designed for projects such as this, already has a large number of machine listening plugins (ugens), and a comprehensive framework for dealing with FFT, audio signals, and much more.
This question about note onset detection contains a lot of information which may be useful to you.
This sounds a huge but very interesting project, good luck to you.
Music transcription means creating music notation from sound (or audio data). While accomplished musicians and especially composers are able to do this, it's an extremely difficult task to do with a machine, and as far as i know, there has been little success so far - mostly academic experiments.
Basically, to recognize notes, you want to know where they start, where they end, and what is their pitch. Fourier transform is the most basic way to turn time domain (audio data) to frequency domain (pitches) - in principle. In practice, musical instruments generate lots of harmonics (overtones) and if we have polyphony (many F0s) added, it's a mess.
You could try feeding something like 50 millisecond sequential slices of the audio data to the FFT. This way you would get the spectrum of each slice, then detect the strongest peaks in each slice, and infer the rhythm from what happens between successive slices.
Sorry, I couldn't help much... But just wanted to point out that what you're trying to do is extremely difficult, seriously. Perhaps you should start from something simpler, like detecting one-note sine wave melodies. Good luck!
For detecting the fundamental frequency of the melody in polyphonic music you can try out the MELODIA vamp plug-in (non-commercial use only): http://mtg.upf.edu/technologies/melodia
If you want to implement a melody extraction algorithm yourself you're going to have to check out the current state-of-the-art in research, a good place to start might be the MIREX melody extraction annual evaluation campaign: http://www.music-ir.org/mirex/wiki/Audio_Melody_Extraction
That, or just google "melody extraction" ;)
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 6 years ago.
Improve this question
Need an algorithm (or a set of good ones to compare for various input data) that will reduce the noise level of a voice audio signal without distorting the signal appreciably using Java.
Input is an audio signal that includes a voice along with some background noises. The noise varies over the course of the recording. There definitely exists ways to remove noise like this, developed for voice recognition and movie making.
The desired output is a minimally distorted voice signal with the background distractions minimally audible to the human ear. The quantitative criteria are minimized
Signal to noise ratio and
Total harmonic distortion.
You are looking for adaptive noise removal and possibly a variety that adapts to changing noise conditions over time within the same stream or file.
Older approaches include:
Remove frequencies containing the majority of noise using bandpass and/or notch filters (which only works well if the desired signal and the noise band do not intersect)
Drop the noise level at points between words, notes, or other audio events (in the dead space) as does the Dolby noise reduction system scheme
Drop the noise floor across an entire file using a Hamming window or other window in conjunction with an FFT library
Hand edit sections of an audio track in programs like Cakewalk or its
competitors
These methods have been found less than desirable when trying to clean up a larger file or multiple files or in real time applications such as real time voice recognition or telephony.
One of the Java programs that I have not personally tried his here. Even though it has some level of automation, it is an LSE (least squares estimator) which works across a block of data but is not suitable for continuous operation or an audio file with changing unwanted noise conditions. (It's not as adaptive as one might hope.)
The solution that I found after much investigation and now use all the time is not written in Java. It is a MATLAB program that can also run in open source Octave with minimal modification. I started porting it to C++ but ran out of time to finish that.
The class of algorithm it implements is called MMSE (Noise Reduction Using Minimum Mean Square Estimators). The MATLAB version has been refined several times by Dr. Hendricks, concluding with the the 2010 version.
I've compared it with its competitors for both dialog and music and it equals or excels beyond the others in all cases I've tried. (I have no affiliation with Dr. Hendricks or MATLAB other than I like the results I've been getting from his implementation on that platform.)
I am working on a personal project. Basically I have a collection of small sound clips, like a clap or a beep noise. I want to create a program that listens for the sounds via a mic or some form of audio input, and when I play sound clip it should identify that clip.
I have tried looking into this myself and have found this article.
http://www.redcode.nl/blog/2010/06/creating-shazam-in-java/
I tried replicating it, but I have found that it doesn't work as expected. I am guessing the sound clips I am using to create my hash from are too small to create enough values to compare.
Wondering if there any well know programs or algorithms that are capable of doing this.
Dan Ellis' slides are probably a good start. They explain the principal task of audio fingerprinting and the two best known approaches:
The Shazam algorithm by A. Wang (paper)
The Philips (now Gracenote) algorithm by Haitsma/Kalker (paper)
As you have already tried the landmark (Shazam) approach, perhaps it's worth your time to fiddle around with the stream-based approach. Since your queries are very short, you might also want to tweak the analysis frame length and overlap. Shorter frames and greater overlap may improve your results for very short samples. If you want to delve even deeper into the Haitsma/Kalker algorithm, you might also be interested in this unfortunately paywalled paper (by me).
i was wondering if anyone has knowledge on the recontruction of 3D objects from live video feed. Does any have any java based examples or papers JAVA based that i could be linked to as i have read up on algorithm's used to produce such 3d objects. If possible i would like to construct something such as the program demostrated in the link provided below.
Currently my program logs live video feed.
http://www.youtube.com/watch?v=brkHE517vpo&feature=related
3D reconstruction of an object from a single point of view is not really possible. You have two basic alternatives: a) To have a stereo camera system capturing the object, b) To have only one camera, but rotating the object (so you will have different points of view of the object), like the one in the video. This is a basic concept related with epipolar geometry.
There are other alternatives, but more intrusive. Some time ago I've been working on a 3D scanner based on a single camera and a laser beam.
For this, I used OpenCV which is C++ code, but now I think there are ports for Java. Have in mind that 3D reconstruction is not an easy task, and the resulting app. will have to be largely parametrized to achieve good results.
This isn't a solved problem - certain techniques can do it to a certain degree under the right conditions. For example, the linked video shows a fairly simple flat-faced object being analysed while moving slowly under relatively even lighting conditions.
The effectiveness of such techniques can also be considerably improved if you can get a second (stereo vision) video feed.
But you are unlikely to get it to work for general video feeds. Problem such as uneven lighting, objects moving in front of the camera, fast motion, focus issues etc. make the problem extremely hard to solve. The best you can probably hope for is a partial reconstruction which can then be reviewed and manually edited to correct the inevitable mistakes.
JavaCV and related projects are probably the best resource if you want to explore further. But don't get your hopes too high for a magic out-of-the-box solution!
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 8 years ago.
Improve this question
I spent quite some time researching for a library that allows me to compare images to one another in Java.
I didn't really find anything useful, maybe my GoogleSearch-skill isn't high enough so I thought I'd ask you guys if you could point me into a direction of where I could find something like this.
Basically what I want to do is to compare two images with each other and get a value of how much the two are similar. Like a percentage or so.
I hope you guys have something I can use, I wouldn't know how to write something like that myself...
PS: It doesn't necessarily has to be in Java, that's just the environment my app will be running.
You could take a look at two answers on SO itself: this one is about image comparison itself, offering links to stuff in C++ (if I read correctly) while this one offers links to broader approaches, one being in C.
I would suggest starting with the second link since there's links on that discussion that'll lead to implementation code of some relevant techniques which you might be able to "translate" into Java yourself.
That's the best my google skills could do, no Java though - sorry. I hope it's a good starting point!
EDIT:
Here's someone with your problem who wrote his own comparison class in Java. I didn't read the source code though. He expressly states that he couldn't find Java libraries for that purpose either, so that's why he wrote it himself.
Oh, and this question on SO has probably the best links on this, all regarding Java libraries of image processing. Hopefully there's one amongst them that can compare images for similarity.
Ok, last edit:
The Java Image Processing Cookbook shows a Java implementation of a basic algorithm to determine the difference between two pictures. It also has an email to contact the guy who wrote it as well as a host of references. No library though.
EDIT after reading your comment to your question:
Unless you've already checked all of the above links, since what you want seems to be checking whether two images are equal, I would suggest starting with the Java Image Processing Cookbook (since that has an implementation of an algorithm in Java to check for equal images) and the last link to an SO question. Also, check PerceptualImageDiff and the source code of that project (C++); it sounds really nifty - it's apparently supposed to check whether two images look equal to the human visual system.
Just off the top of my head, OpenCV is a great image processing library, but it might be overkill if you just want to compare images. If that's the case, I'd go with ImageJ.
Someone already asked how to do this using OpenCV here.
I'd use C++ for this, but if you must use Java, there is a project which made a Java wrapper for OpenCV, here.
I used the class in this link to compare two product images, and the results were cool. It's not very hard to implement it just to be used for comparing two images, you just need to delete the lines of JAI and Swing and such. It resizes images to 300x300 and returns a difference value such as "1234". The maximum difference value is near "11041", it's stated in the link. Doing a division, you can simply get the percentage. If interested I can post the modified code here later.
The results were cool, but I still got "digital camera photos", detected to be similar to "TV photos". So, I used ImageJ to detect edges in the picture. Using the detect edges operation, ImageJ converts the image into a edge detected greyform image. Than I put the two edge-detected images in the same comparator and multiplied the both values. The results got even more accurate.
Getting the edge-detected form of the images
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 4 years ago.
Improve this question
Now that the oligopole of market data providers successfully killed OpenQuant, does any alternative to proprietary and expensive subscriptions for realtime market data subsist?
Ideally I would like to be able to monitor tick by tick securities from the NYSE, NASDAQ and AMEX (about 6000 symbols).
Most vendors put a limit of 500 symbols watchable at the same time, this is unacceptable to me, even if one can imagine a rotation among the 500 symbols ie. making windows of 5 sec. of effective observation out of each minute for every symbol.
Currently I'm doing this by a Java thread pool calling Google Finance, but this is unsatisfactory for several reasons, one being that Google doesn't return the volume traded, but the main one being that Google promptly is killing bots attempting to take advantage of this service ;-)
Any hint much appreciated,
Cheers
I think you'll find all you need to know by looking at this question: source of historical stock data
I don't know of any free data feeds other than Yahoo!, but it doesn't offer tick-by-tick data, it only offers 1 minute intervals with a 15 minute delay. If you want to use an already existing tool to download the historical data, then I would recommend EclipseTrader. It only saves the Open, Close, High, Low, and Volume.
(source: divbyzero.com)
You can write your own data scraper with very little effort. I've written an article on downloading real-time data from yahoo on my blog, but it's in C#. If you're familiar with C# then you'll be able to translate the action in Java pretty quickly. If you write your own data scraper then you can get pretty much ANYTHING that Yahoo! shows on their web site: Bid, Ask, Dividend Share, Earnings Share, Day's High, Day's Low, etc, etc, etc.
If you don't know C# then don't worry, it's REALLY simple: Yahoo allows you to download CSV files with quotes just by modifying a URL. You can find out everything about the URL and the tags that are used on yahoo here: http://www.gummy-stuff.org/Yahoo-data.htm
Here are the basic steps you need to follow:
Construct a URL for the symbol or multiple symbols of your choice.
Add the tags which you're interested in downloading (Open, Close, Volume, Beta, 52 week high, etc, etc.).
Create a URLConnection with the URL you just constructed.
Use a BufferedReader to read the CSV file that is returned from the connection stream.
Your CSV will have the following format:
Each row is a different symbol.
Each column is a different tag.
Open a TDAmeritrade account and you will have free access to ThinkOrSwim real time trading and quotes platform. Live trading is real time and paper trading is delayed 15 minutes. I forget what the minimum required is to open a TDAmeritrade account but you can go to TDAMeritrade.com or thinkorswim.com to check them out.
Intrinio has a bunch of feeds with free and paid tiers. Essentially you only have to pay for what you need as opposed to the bigger data suppliers. Intrinio focuses on data quality and caters to developers as well, so I think it'd be a great option for you.
full disclosure - I work at Intrinio as a developer
There's a handy function in Google Sheets (ImportHTML) which I've been using for a while to reasonable effect.
For example -
=ImportHTML("http://www.bloomberg.com/markets/commodities/futures/metals/","table",1),5,3) returns the EUR Gold spot price.
It works with Yahoo too, so =Index(ImportHTML("http://finance.yahoo.com/q?s=DX-Y.NYB","table",0),2,2) returns the DXY.
The data updates with some small delay but it's usable.