is there a way to detect either a Video or Audio track present in a raw MP4 data?
I have downloaded a raw MP4 data (with some size) from a URL with a range request. It is not a full MP4 data.
I am using MP4Parser to do this job.
What I have tried is I am trying to build a Movie object from the above MP4 data. But it throws exception saying that there are no tracks in it.
Any idea would be much appreciated how to construct Movie object from a raw MP4 data.
Thanks
without the original moov box, the file is pretty much worthless. It is possible to locate the frames in the mdat and reconstruct a new moov with some forensic analysis, but its almost certainly not work the effort or cost dot so.
Related
Is there any native/existing/manual way to remove all the metadata of a video file in Android/Java/Kotlin from a (content) URI?
For photos, we can still use ExifInterface and remove the metadata we want by using setAttribute and passing null to delete the given/desired values, however this doesn't work for videos because an mp4 video has it's own metadata format. If there is no native Android class that can do this, is there any algorithm or reference that can be referred in order to implement this?
You can use mp4parser library. Typical tasks for the MP4 Parser are:
Muxing audio/video into an MP4 file
Append recordings that use the same encode settings
Adding/Changing metadata
Shorten recordings by omitting frames
The MetaDataInsert example shows how to write metadata.
I am trying to read the audio data using MIC, and I am able to successfully read and save it to a file (wav format) returned by the AudioRecord class.
Now the real problem is that the file I am creating is too big. say Audio with duration of 5 minutes is taking upto 25MB.
Can anyone suggest me how to reduce the size. I am open to other file formats as well.
Thanks in advance.
WAV is expensive. Try MP3. See the code - https://github.com/yhirano/Mp3VoiceRecorderSampleForAndroid
Mp3 Encoder: http://www.tritonus.org/plugins.html
Try using Android's MediaMuxer. It is only supported on API 18+
I have tons of ripped .wav files (I'm ready to convert them into flacs if it's easier) which details I want to insert in a MySQL database. When I right click the .wav files in Windows Explorer (not the browser) and select Properties -> Details I can see some details about the song. For example the artist, genre and duration. How can I read and edit these details in Java?
To get durration information, see this link: Java - reading, manipulating and writing WAV files
Essentially, a WAV file is broken up into chunks, which either contain audio data, or describe the audio data in some way, or provide information about it. If the reader doesn't understand one of those chunks it is able to skip it, which allows placing a lot of different kinds of information in the file. One of those chunks contains information like the samplerate, number of channels and total number of sample frames, from which you can calculate the length.
For artist, genre and so on... well there's no standard chunk for that, so if that's really in the file, and not in the windows db somewhere, it's probably stored in ID3 tags embedded in the WAV. I don't know for sure what the chunkID is for ID3, but it's probably "id3 ", or "ID3 " (including the space). You coud probably figure this out by searching for strings of length 4 or more in the file -- usually data chunks are in the beginning and audio is at the end. (on unix/macos I would use the "strings" command, maybe with "head") ID3 tags are standard for MP3, and you can figure out how to parse them by googling. To get to them, you'll need to understand WAV files first, at least enough to know what chunks are, chunkIds, how to skip chunks you don't care about, and so on.
I don't know of a library that will read ID3 tags in WAV files in Java, so you'll either have to write one, or wrap one written in another language. I suspect libsndfile will work, but it doesn't have an MP3 reader, so maybe not. You could also try SOX. You can also check out http://javamusictag.sourceforge.net/ which I've never used, but it came up in a search.
good luck!
I ended up converting them into flac files and using JAudiotagger. Thanks for the responses, this time I ended up this way.
http://www.jthink.net/jaudiotagger/
I need to split mpeg4 video stream (actually from android video camera) to send it through RTP.
The specification is little large for quick reference.
I wonder if there any example/open source code for mpeg4 packetization?
Thanks for any help !
Mpeg4 file format is also called ISO/IEC 14496-14. Google it any you will find specifications.
However, what you are trying to do (RTP publisher) will be hard for the following reasons:
Mpeg4 has header at the end of the file. Which means header will be written out only when video stream is finished. Since you want to do real time video streaming you will need to guess where audio and video packets start/end. This will not be the same on all Android devices as they might use different video sizes and codec parameters. So your code will be device-dependent and you'll need to support and test many different devices.
Some devices do not flush video data to file in regular intervals. Some only flush once a minute or so. This will break your real-time stream.
There is no example code. I know because I looked. There are a few companies that do something similar, but mainly they skip RTP. Instead they progressively upload the file to their own server and then implement video/audio stream "chopping" and then insert it into their video/transcoder backend. I used to work for one of those companies and that's how we did it. AFAIK competition took similar approaches. The upside is that all complexity is on server an you do not need to update clients when something breaks or new devices arrive on the market.
Are there libraries out there that can convert data (text files, etc) to sound and back to the original data?
The sound can be transmitted any medium I wish, whether radio, etc. I just need to store data in sound files.
Scenario:
step1: Convert a .docx file with embedded images to .wav.
step2: Send over a radio wave.
step3: Convert this .wav back to the .docx file with the embedded images.
This concept can be applied to any data.
Technology:
.net or java
I think the medium is important, as are other factors such as the size of the files and the transmission time available. A simple algorithm would be to convert your files to text (UUENCODE should do that trick) then convert to morse code : http://www.codeproject.com/KB/vb/morsecode.aspx
Morse gives you a simple alphabet able to survive transmission over a fairly noisy radio channel.
If your carrier is cleaner a conversion of your UUEncoded file into a series of frequencies one per character would probably also work, and be easy enough to decode at the other end, Frequency Analyzer in C#
You could try to use the magnetic card technology for your files, I'm also trying to do this on android.
Any data can be converted to byte into a string of characters it very possible with java and android.
then use the Encoding mechanism of Magnetic Cards API to encode the string to sound. Then you can just use the vice versa, convert the sound into string convert string into byte and save the data. It's just it takes time to convert both ways but it is feasible, I'm trying to do this so that any one with unlimited voice connection can transfer files or in the future browse the internet just through calling the other number. I hope I gave you some idea.
The problem is that the data in a word document doesn't necessarily make decent sound. If you pick a 1.8kHz carrier and use the binary contents of the word document to modulate the volume or the frequency (AM or FM) the result will be messy and hardly to decode.
But if you save the document as a bitmap, you can use the pixel values to modulate the volume of the carrier wave.
We've been sending pictures (not just black/white but greyscale and color (three different separations of the image, r, g and b) over phonelines using this method for many years before modems and the internet took off.
The fun part is that you can broadcast data this way. The sound can be received by more than one receiver at the same time. There's no error correction, but as you deal with visual data, you don't have to worry about a few pixels getting lost. It's similar to old fax protocols.
Does the audio file need to be convertible using lossy compressors (MP3 etc.)? If not, you can just add a WAV container around any binary data and you'll be fine. Otherwise it gets more difficult, and you need to ensure that the audio is audible (in a reasonable frequency range when played) and be tolerant enough on the frequency detection to match the output of lossy codecs.
Best way is to convert the audio file into binary and store in a file type you specify.
Try out the AudioInputStream Class in Java
To give what I think is a better response to all of the above, have a look at packet radio and the various bits that correspond to it AX.25 is a good example and there are a number of implementations for it. POCSAG is also another good implementation. Both have libraries available for many different languages and have been around for quite a long time.
Other example include things like WEFAX (weather fax), HFFax, SSTV (slow scan tv), etc.
You can think of them all as being similar to the old school phone line modem type encoders and decoders that run around the 300-2400baud