My goal is to obtain the rpm of powerballs by analysing sound generated during ball rotation. In google play, several application available to calculate rpm and frequency by sound but each and every app failed to find out rpm of power balls. I searched on google and found that "To calculate frequency from sound, we have to use FFT Algo".
what is power ball?
Powerball rotate and generate sound.
"Powerball® generates resistance which is directly in proportion to the effort expended by the user".
class RecorderThread1 extends Thread {
private static final String TAG = RecorderThread1.class.getSimpleName();
public boolean recording; // variable to start or stop recording
public int frequency; // the public variable that contains the frequency
private PlaceholderFragment placeHolder;
private Handler handler;
private long avgFrequency;
private int avg = 0;
private int sampelRateHz = 8000;
// value "heard", it is updated continually while
// the thread is running.
public RecorderThread1(PlaceholderFragment placeHolder) {
this.placeHolder = placeHolder;
}
#Override
public void run() {
AudioRecord recorder;
int numCrossing, p;
short audioData[];
int bufferSize;
bufferSize = AudioRecord.getMinBufferSize(8000,
AudioFormat.CHANNEL_CONFIGURATION_MONO,
AudioFormat.ENCODING_PCM_16BIT) * 3; // get the buffer size to
// use with this audio
// record
recorder = new AudioRecord(AudioSource.MIC, 8000,
AudioFormat.CHANNEL_CONFIGURATION_MONO,
AudioFormat.ENCODING_PCM_16BIT, bufferSize); // instantiate the
// AudioRecorder
recording = true; // variable to use start or stop recording
audioData = new short[bufferSize]; // short array that pcm data is put
// into.
while (recording) { // loop while recording is needed
if (recorder.getState() == android.media.AudioRecord.STATE_INITIALIZED) // check
// to
// see
// if
// the
// recorder
// has
// initialized
// yet.
if (recorder.getRecordingState() == android.media.AudioRecord.RECORDSTATE_STOPPED)
recorder.startRecording(); // check to see if the Recorder
// has stopped or is not
// recording, and make it
// record.
else {
recorder.read(audioData, 0, bufferSize); // read the PCM
// audio data
// into the
// audioData
// array
// Now we need to decode the PCM data using the Zero
// Crossings Method
numCrossing = 0; // initialize your number of zero crossings
// to 0
for (p = 0; p 0 && audioData[p + 1] = 0)
numCrossing++;
if (audioData[p + 1] > 0 && audioData[p + 2] = 0)
numCrossing++;
if (audioData[p + 2] > 0 && audioData[p + 3] = 0)
numCrossing++;
if (audioData[p + 3] > 0 && audioData[p + 4] = 0)
numCrossing++;
}// for p
for (p = (bufferSize / 4) * 4; p 0 && audioData[p + 1] = 0)
numCrossing++;
}
frequency = (8000 / bufferSize) * (numCrossing / 2); // Set
// the audio Frequency to half the number of zero crossings, times the number of samples our buffersize is per second.
avgFrequency = avgFrequency + frequency;
avg++;
Log.d(TAG, " frequency is " + frequency);
if (handler == null) {
handler = new Handler(Looper.getMainLooper());
handler.postDelayed(runnable, 2000);
}
// placeHolder.printFrequency((frequency * 60));
}// else recorder started
} // while recording
The number of zero crossing will not be reliable, higher and lower frequencies will add noise, unless the sound is a pure sinus or square wave.
You can find and FFT algorithm there: FFT library in android Sdk
FFT is computing the power in frequency bands (as a spectrum analyzer would display). You should try to look at the curve image from the powerball sound, and decide which is the best way to handle it (lower frequency peak, higher value peak, ...).
Related
In my app I generate list of sounds with different frequencies, for example 1000Hz, 2000Hz, 4000Hz ... for left and right channel:
private AudioTrack generateTone(Ear ear) {
// fill out the array
int numSamples = SAMPLE_RATE * getDurationInSeconds();
double[] sample = new double[numSamples];
byte[] generatedSnd = new byte[2 * numSamples];
for (int i = 0; i < numSamples; ++i) {
sample[i] = Math.sin(2 * Math.PI * i / (SAMPLE_RATE / getLatestFreqInHz()));
}
// convert to 16 bit pcm sound array
// assumes the sample buffer is normalised.
int idx = 0;
for (final double dVal : sample) {
// scale to maximum amplitude
final short val = (short) ((dVal * 32767));
// in 16 bit wav PCM, first byte is the low order byte
generatedSnd[idx++] = (byte) (val & 0x00ff);
generatedSnd[idx++] = (byte) ((val & 0xff00) >>> 8);
}
AudioTrack audioTrack = new AudioTrack(AudioManager.STREAM_MUSIC,
SAMPLE_RATE, AudioFormat.CHANNEL_OUT_MONO,
AudioFormat.ENCODING_PCM_16BIT, numSamples,
AudioTrack.MODE_STATIC);
audioTrack.write(generatedSnd, 0, generatedSnd.length);
int channel;
if (ear == Ear.LEFT) {
audioTrack.setStereoVolume(1.0f, 0.0f);
} else if (ear == Ear.RIGHT) {
audioTrack.setStereoVolume(0.0f, 1.0f);
}
return audioTrack;
}
I use this code to control volume:
audioManager = (AudioManager) context.getApplicationContext().getSystemService(Context.AUDIO_SERVICE);
MAX_VOLUME = audioManager.getStreamMaxVolume(STREAM_MUSIC);
MIN_VOLUME = audioManager.getStreamMinVolume(STREAM_MUSIC);
APP_DEFAULT_VOLUME = (MAX_VOLUME + MIN_VOLUME) /2 ;
#Override
public void setToAppDefaultVolume() {
Timber.tag(TAG).d("Resetting to app default sound volume.");
audioManager.setStreamVolume(STREAM_MUSIC, APP_DEFAULT_VOLUME, 0);
}
#Override
public void increaseVolume() {
if (!isReachedMaxVolume()) {
Timber.tag(TAG).d("Increasing device sound volume from %d to %d", currentVolumeLevel(), currentVolumeLevel() + STEP);
audioManager.setStreamVolume(STREAM_MUSIC, currentVolumeLevel() + STEP, 0);
} else {
Timber.tag(TAG).d("Reached the maximum device volume.");
}
}
#Override
public void decreaseVolume() {
if (!isReachedMinVolume()) {
Timber.tag(TAG).d("Decreasing device sound volume from %d to %d", currentVolumeLevel(), currentVolumeLevel() - STEP);
audioManager.setStreamVolume(STREAM_MUSIC, currentVolumeLevel() - STEP, 0);
} else {
Timber.tag(TAG).d("Reached the preferred minimum volume");
}
}
I start playing each tone (frequency) with APP_DEFAULT_VOLUME and gradually increase tone. When user confirm he/she heard specific tone with specific volume, I want to calculate it's amplitude and log it for latter so I can review user hearing ...
But i have no clue how to do so!
All solutions I found was about reading data from microphone and calculate the amplitude to visualize it on screen ...
My scenario is much simpler. I have device volume recorded, frequency is fixed and recorded, and Audio is generated dynamically and is not a file.
Will anyone help me with this scenario ?
Thanks in advance.
I want to play a generated sound that is shorter than 1 second. However, the minBufferSize of the AudioTrack always seems to be 1 second or longer. On some devices I can set the bufferSize smaller than the value evaluated with AudioTrack.getMinBufferSize, however this is not possible on all devices. I'd like to know wether it's possible to generate a shorter sound for the AudioTrack. I'm currently using this code (it contains some smoothing, because I'm getting constantly new frequences):
int buffSize = AudioTrack.getMinBufferSize(sampleRate,
AudioFormat.CHANNEL_OUT_MONO,
AudioFormat.ENCODING_PCM_16BIT);
AudioTrack audioTrack = new AudioTrack(AudioManager.STREAM_MUSIC, sampleRate,
AudioFormat.CHANNEL_OUT_MONO,
AudioFormat.ENCODING_PCM_16BIT, buffSize,
AudioTrack.MODE_STREAM);
short samples[] = new short[buffSize];
int amp = 10000;
double twopi = 8. * Math.atan(1.);
double phase = 0.0;
audioTrack.play();
double currentFrequency = getFrequency();
double smoothing = 300;
double deltaTime = buffSize / 500;
while (playing && PreferenceManager.getDefaultSharedPreferences(
MainActivity.this).getBoolean("effect_switch", true))
{
double newFrequency = getFrequency();
for (int i = 0; i < buffSize; i++)
{
currentFrequency += deltaTime * (newFrequency - currentFrequency) / smoothing;
samples[i] = (short) (amp * Math.sin(phase));
phase += twopi * currentFrequency / sampleRate;
}
audioTrack.write(samples, 0, buffSize);
}
audioTrack.stop();
audioTrack.release();
In fact, I want the sounds to be updated more frequently, which is the reason for me needing shorter samples.
I think I have a solution for you. Since my min buffer seems to be much smaller than 1 sec, I simulated your problem by loading a buffer with 5 sec of data but only play 0.5 sec of it immediately followed by another frequency. This tone I also created 5 sec of data but only played 0.5 sec & repeated this for several tones. It all works for me.
Also, since I jammed this into a current project I'm working on, it's difficult for me to just cut and paste my code. While I've tested my solution, what I've posted here is not tested exactly as written. Some of it is cut & paste, some pseudocode.
The key feature is using the OnPlaybackPositionUpdateListener.
private AudioTrack.OnPlaybackPositionUpdateListener audioTrackListener = new AudioTrack.OnPlaybackPositionUpdateListener() {
#Override
public void onMarkerReached(AudioTrack audioTrack) {
int marker = audioTrack.getNotificationMarkerPosition();
// I just used 8 tones of 0.5 sec each to determine when to stop but you could make
// the condition based on a button click or whatever is best for you
if(marker < MAX_FRAME_POSITION) {
audioTrack.pause();
newSamples();
audioTrack.play();
} else {
audioTrack.stop();
}
audioTrack.setNotificationMarkerPosition(marker + FRAME_MARKER);
Log.d(TAG, "MarkerReached");
}
#Override
public void onPeriodicNotification(AudioTrack audioTrack) {
int position = audioTrack.getPlaybackHeadPosition();
if(position < MAX_FRAME_POSITION) {
audioTrack.pause();
newSamples();
audioTrack.play();
} else {
audioTrack.stop();
}
Log.d(TAG, "PeriodNotification");
}
};
Then
audioTrack.setPlaybackPositionUpdateListener(AudioTrackListener);
I used the marker (which has to be re-initialized repeatedly) for my tests...
audioTrack.setNotificationMarkerPosition(MARKER_FRAMES);
but you should be able to use the periodic notification too.
audioTrack.setPositionNotificationPeriod(PERIODIC_FRAMES);
And the newSamples() method called from the listener
public void newSamples() {
/*
* generate buffer, I'm doing similar to you, without the smoothing
*/
// AudioTrack write is a blocking operation so I've moved it off to it's own Thread.
// Could also be done with an AsyncTask.
Thread thread = new Thread(writeSamples);
thread.start();
}
private Runnable writeSamples = new Runnable() {
#Override
public void run() {
audioTrack.write(samples, 0, buffSize);
}
};
I am making an app where I need to read continues stream of sound which is sent in the form of an byte array. The server side records sound like this (based on an example here on SO):
// Get the minimum buffer size required for the successful creation of an AudioRecord object.
int bufferSizeInBytes = AudioRecord.getMinBufferSize(RECORDER_SAMPLERATE, RECORDER_CHANNELS,
RECORDER_AUDIO_ENCODING);
bufferSizeInBytes = 30000;
// Initialize Audio Recorder.
_audio_recorder = new AudioRecord(MediaRecorder.AudioSource.MIC, RECORDER_SAMPLERATE,
RECORDER_CHANNELS, RECORDER_AUDIO_ENCODING, bufferSizeInBytes);
// Start Recording.
_audio_recorder.startRecording();
int numberOfReadBytes = 0;
byte audioBuffer[] = new byte[bufferSizeInBytes];
boolean recording = false;
float tempFloatBuffer[] = new float[3];
int tempIndex = 0;
byte totalByteBuffer[] = new byte[60 * 44100 * 2];
while (true)
{
float totalAbsValue = 0.0f;
short sample = 0;
numberOfReadBytes = _audio_recorder.read(audioBuffer, 0, bufferSizeInBytes);
for (int i = 0; i < bufferSizeInBytes; i += 2)
{
sample = (short) ((audioBuffer[i]) | audioBuffer[i + 1] << 8);
totalAbsValue += Math.abs(sample) / (numberOfReadBytes / 2);
}
tempFloatBuffer[tempIndex % 3] = totalAbsValue;
float temp = 0.0f;
for (int i = 0; i < 3; ++i)
temp += tempFloatBuffer[i];
if ((temp >= 0 && temp <= _sensitivity) && recording == false)
{
Log.i("TAG", "1");
tempIndex++;
continue;
}
if (temp > _sensitivity && recording == false)
{
Log.i("TAG", "2");
recording = true;
}
if(temp < _sensitivity && recording == true)
{
recording = false;
continue;
}
for (int i = 0; i < numberOfReadBytes; i++)
totalByteBuffer[i] = audioBuffer[i];
if (prepare_sound(totalByteBuffer, numberOfReadBytes))
{
totalByteBuffer = new byte[60 * 44100 * 2];
tempIndex++;
}
}
The example this is taken from is recording sound and saves it to a file when there is no more sound to record. My goal on the other hand is to record sound when there is sound and send this sound on the fly when still recording. Hence, I want to send sounds right a way and not store it to a file when there is no more sound to record. So far I am taking the byte[] with data and stores it in an object an sends it to a client using ObjectOutputStream. The client will then create a temp sound file and play it using MediaPlayer. But I feel that this is not the most effect way to achieve my goal. So, is there any more efficient way to do this with respect to send an continues stream of data as media player does not support playing pure byte[] of data?
Thanks for any help and tips!
Found out that the best solutions for me is to record the sound and when the buffer is full it is sent to the client side. The client then uses an AudioTrack instance to play the byte[] that contains the data like this:
public void onSoundReceived(byte[] sound)
{
_audio_input_stream.write(sound, 0, sound.length);
}
This also makes the sound more "non-lagging" as this is not a MediaPlayer instance which will stop the sound after each time the data was done playing.
I have a problem to generate smooth sinus wave.
I've done it few years ago on C++ and everything worked perfect. Now I am trying to do this using AudioTrack and I do not know what is wrong.
This is my test case:
I want to produce for five second a sinus wave which is smooth (no crack etc.). For one second I generate 44100 samples and divided it on couple of buffer with size 8192 (probably this is the reason of cracks, but how can I fix it, without giving bigger size of buffer).
Unfortunatelly using my code the sound is not smooth and instead of 5 second it takes about 1 second. I would be gratefull for any help.
Please let me now if this piece of code is not enough.
class Constants:
//<---
public final static int SAMPLING = 44100;
public final static int DEFAULT_GEN_DURATION = 1000;
public final static int DEFAULT_NUM_SAMPLES = DEFAULT_GEN_DURATION * SAMPLING / 1000; //44100 per second
public final static int DEFAULT_BUFFER_SIZE = 8192;
//--->
//preparing buffers to play;
Buffer buffer = new Buffer();
short[] buffer_values = new short[Constants.DEFAULT_BUFFER_SIZE];
float[] samples = new float[Constants.DEFAULT_BUFFER_SIZE];
float d = (float) (( Constants.FREQUENCIES[index] * 2 * Math.PI ) / Constants.SAMPLING);
int numSamples = Constants.DEFAULT_NUM_SAMPLES; //44100 per second - for test
float x = 0;
int index_in_buffer = 0;
for(int i = 0; i < numSamples; i++){
if(index_in_buffer >= Constants.DEFAULT_BUFFER_SIZE - 1){
buffer.setBufferShort(buffer_values);
buffer.setBufferSizeShort(index_in_buffer);
queue_with_data_AL.add(buffer); //add buffer to queue
buffer_values = new short[Constants.DEFAULT_BUFFER_SIZE];
samples = new float[Constants.DEFAULT_BUFFER_SIZE];
index_in_buffer = 0;
}
samples[index_in_buffer] = (float) Math.sin(x);
buffer_values[index_in_buffer] = (short) (samples[index_in_buffer] * Short.MAX_VALUE);
x += d;
index_in_buffer++;
}
buffer.setBufferShort(buffer_values);
buffer.setBufferSizeShort(index_in_buffer+1);
queue_with_data_AL.add(buffer);
index_in_buffer = 0;
}
//class AudioPlayer
public AudioPlayer(int sampleRate) { //44100
int minSize = AudioTrack.getMinBufferSize(sampleRate,
AudioFormat.CHANNEL_OUT_MONO, AudioFormat.ENCODING_PCM_16BIT);
audiotrack = new AudioTrack(AudioManager.STREAM_MUSIC, sampleRate,
AudioFormat.CHANNEL_OUT_MONO, AudioFormat.ENCODING_PCM_16BIT,
minSize, AudioTrack.MODE_STREAM);
}
public void play(byte[] audioData, int sizeOfBuffer) {
audiotrack.write(audioData, 0, sizeOfBuffer);
}
public void start() {
if (state == Constants.STOP_STATE) {
state = Constants.START_STATE;
int startLength = 0;
while (state == Constants.START_STATE) {
Buffer buffer = getBufferFromQueueAL(); //getting buffer from prepared list
if (buffer != null) {
short[] data = buffer.getBufferShort();
int size_of_data = buffer.getBufferSizeShort();
if (data != null) {
int len = audiotrack.write(data, 0, size_of_data);
if (startLength == 0) {
audiotrack.play();
}
startLength += len;
} else {
break;
}
} else {
MessagesLog.e(TAG, "get null data");
break;
}
}
if (audiotrack != null) {
audiotrack.pause();
audiotrack.flush();
audiotrack.stop();
}
}
}
You are playing only 1 second because 44100 samples at a samplerate of 44100Hz result in exactly 1 second of sound.
You have to generate 5 times more samples if you want to play 5 seconds of sound (e.g. multiply DEFAULT_NUM_SAMPLES by 5) in your code.
I've found the solution by myself. After adding Buffer to queue_with_data_AL I've forgotten create new instance of Buffer object. So in queue was couple of buffer with the same instance, hence sinus wave were not continuous.
Thanks if someone was trying to solve my problem. Unfortunatelly it was my programming mistake.
Best regards.
I am using android platform, from the following reference question I come to know that using AudioRecord class which returns raw data I can filter range of audio frequency depends upon my need but for that I will need algorithm, can somebody please help me out to find algorithm to filter range b/w 14,400 bph and 16,200 bph.
I tried "JTransform" but i don't know can I achieve this with JTransform or not ? Currently I am using "jfftpack" to display visual effects which works very well but i can't achieve audio filter using this.
Reference here
help appreciated Thanks in advance.
Following is my code as i mentioned above i am using "jfftpack" library to display you may find this library reference in the code please don't get confuse with that
private class RecordAudio extends AsyncTask<Void, double[], Void> {
#Override
protected Void doInBackground(Void... params) {
try {
final AudioRecord audioRecord = findAudioRecord();
if(audioRecord == null){
return null;
}
final short[] buffer = new short[blockSize];
final double[] toTransform = new double[blockSize];
audioRecord.startRecording();
while (started) {
final int bufferReadResult = audioRecord.read(buffer, 0, blockSize);
for (int i = 0; i < blockSize && i < bufferReadResult; i++) {
toTransform[i] = (double) buffer[i] / 32768.0; // signed 16 bit
}
transformer.ft(toTransform);
publishProgress(toTransform);
}
audioRecord.stop();
audioRecord.release();
} catch (Throwable t) {
Log.e("AudioRecord", "Recording Failed");
}
return null;
/**
* #param toTransform
*/
protected void onProgressUpdate(double[]... toTransform) {
canvas.drawColor(Color.BLACK);
for (int i = 0; i < toTransform[0].length; i++) {
int x = i;
int downy = (int) (100 - (toTransform[0][i] * 10));
int upy = 100;
canvas.drawLine(x, downy, x, upy, paint);
}
imageView.invalidate();
}
There are a lot of tiny details in this process that can potentially hang you up here. This code isn't tested and I don't do audio filtering very often so you should be extremely suspicious here. This is the basic process you would take for filtering audio:
Get audio buffer
Possible audio buffer conversion (byte to float)
(optional) Apply windowing function i.e. Hanning
Take the FFT
Filter frequencies
Take inverse FFT
I'm assuming you have some basic knowledge of Android and audio recording so will cover steps 4-6 here.
//it is assumed that a float array audioBuffer exists with even length = to
//the capture size of your audio buffer
//The size of the FFT will be the size of your audioBuffer / 2
int FFT_SIZE = bufferSize / 2;
FloatFFT_1D mFFT = new FloatFFT_1D(FFT_SIZE); //this is a jTransforms type
//Take the FFT
mFFT.realForward(audioBuffer);
//The first 1/2 of audioBuffer now contains bins that represent the frequency
//of your wave, in a way. To get the actual frequency from the bin:
//frequency_of_bin = bin_index * sample_rate / FFT_SIZE
//assuming the length of audioBuffer is even, the real and imaginary parts will be
//stored as follows
//audioBuffer[2*k] = Re[k], 0<=k<n/2
//audioBuffer[2*k+1] = Im[k], 0<k<n/2
//Define the frequencies of interest
float freqMin = 14400;
float freqMax = 16200;
//Loop through the fft bins and filter frequencies
for(int fftBin = 0; fftBin < FFT_SIZE; fftBin++){
//Calculate the frequency of this bin assuming a sampling rate of 44,100 Hz
float frequency = (float)fftBin * 44100F / (float)FFT_SIZE;
//Now filter the audio, I'm assuming you wanted to keep the
//frequencies of interest rather than discard them.
if(frequency < freqMin || frequency > freqMax){
//Calculate the index where the real and imaginary parts are stored
int real = 2 * fftBin;
int imaginary = 2 * fftBin + 1;
//zero out this frequency
audioBuffer[real] = 0;
audioBuffer[imaginary] = 0;
}
}
//Take the inverse FFT to convert signal from frequency to time domain
mFFT.realInverse(audioBuffer, false);