Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/jomarkow/Audio-to-MIDI
Real time Audio to MIDI converter. Convert your sound into an editable music document.
https://github.com/jomarkow/Audio-to-MIDI
arduino audio audio-processing cpp fft midi music stft teensy teensy40
Last synced: 2 months ago
JSON representation
Real time Audio to MIDI converter. Convert your sound into an editable music document.
- Host: GitHub
- URL: https://github.com/jomarkow/Audio-to-MIDI
- Owner: jomarkow
- Created: 2023-06-06T18:00:45.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2023-08-17T14:40:06.000Z (over 1 year ago)
- Last Synced: 2024-08-01T02:27:50.143Z (5 months ago)
- Topics: arduino, audio, audio-processing, cpp, fft, midi, music, stft, teensy, teensy40
- Language: C++
- Homepage:
- Size: 1.03 MB
- Stars: 4
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# AUDIO TO MIDI in real time
#### Convert your sound into an editable music document.
### Index:
* [Description](#description)
* [Software](#software)
* [Electronic Filter](#filter)
* [Simulation](#simulation)
* [BOM](#bom)
* [Colaborate](#colab)
The idea of this project originates from an interest on music and trying to find a way to make it easier to record and edit it at the moment. Since me and my teammates enjoy both music and electronics, we decided to go this way and create this project.
The objective of this project is to make the user able to convert real sounds, whether of instruments or voice, into a MIDI format in real time. With MIDI you can put these notes into a music editor, like Reaper, and saw them the while you are playing.
This was made in team with:
[Joaquin Pimpignano](https://github.com/JoaquinPimpignano),
[Iris Martinez](https://github.com/IrisMartinez05),
[Juan Cruz Martinez]() and
[Violeta Martinez]().
#### Made with:
[](https://www.arduino.cc/)
[](https://www.pjrc.com/store/teensy40.html)
[](https://www.kicad.org/)
The most important part of the project. We decide to do it on a teensy because of his memory and fast processment capacity, making posible to create a Fourier Transform for analize the signal and discover the notes in real time.
It tooks the audio and analize it with FFT and a windowing function for making it faster, it returns the dominant frequency.
```
float Calculate_FFT() {
FFT = arduinoFFT(vReal, vImag, SAMPLES, SAMPLING_FREQ);
FFT.Windowing(FFT_WIN_TYP_HAMMING, FFT_FORWARD);
FFT.Compute(FFT_FORWARD);
FFT.ComplexToMagnitude();
double dominantFrequency = FFT.MajorPeak();
return dominantFrequency;
}
```
Then, it looks this frequency on a defined array with the corresponding notes for each frequency range.
```
float NOTES_FREQ[] = { 103.826, 110.000, 116.541, 123.471, 130.813, 138.591, 146.832, 155.563, 164.814, 174.614,
184.997, 195.998, 207.652, 220.000, 233.082, 246.942, 261.626, 277.183, 293.665, 311.127,
329.628, 349.228, 369.994, 391.995, 415.305, 440.000, 466.164, 493.883, 523.251, 554.365,
587.330, 622.254, 659.255, 698.456, 739.989, 783.991, 830.609, 880.000, 932.328, 987.767 };
}
```
[](https://postimg.cc/WF9YrsX7)
```
#define MAX_NOTE_FREQ 987.767
#define MIN_NOTE_FREQ 103.826
#define OCTAVES_FITTING 55
float searchForNote(float freq) {
if (freq > MAX_NOTE_FREQ || freq < MIN_NOTE_FREQ) return 0;
for (int i = 1; i < NOTES_LENGHT - 1; i++) {
if (freq > NOTES_FREQ[i - 1] && freq < NOTES_FREQ[i]) {
if (freq - NOTES_FREQ[i - 1] < NOTES_FREQ[i] - freq) return i + OCTAVES_FITTING;
else return i + OCTAVES_FITTING + 1;
}
}
}
```
The project has also an electronic board who contains a circuit with:
* Amplifier, for make the microphone signal stronger
* Band pass filter between 100hz and 1Khz
* Clamper circuit, for elevate the signal to the positive field
[](https://postimg.cc/Kkj0rdKr)
Board designed on kicad 7.0:
[](https://postimg.cc/tnt6rN8X)
We simulate the circuit on Proteus, here the results:
[](https://postimg.cc/p5xpttDK)
As you could see, as more amplitude on the input, the filter gets better on cutting the frequencies out of our range, but in general, it gives us a good result.
On "simulation" folder you will find the proteus file for simulate it yourself.
Component|Quantity|
|:-------------|:-------------|
|Teensy 4.0|1|
|Microfono Electret|1|
|Pot 20K ohm|1|
|R de 10K ohm|5|
|R de 100K ohm|1|
|R de 100 ohm|1|
|R de 1,5K ohm|1|
|R de 1M ohm|1|
|Cap Elec 10u|1|
|Cap Elec 100n|1|
|Cap 100n|1|
|Cap 1u|1|
|Cap 10n|1|
|Cap 33u|1|
|Lm358n|2|
|Lm7805|2|
|2N4001|1|
|L7806|1|
|ld1117|1|
We see in this project a great potencial for implementing diferent technologies in future versions, such as an IA for better sound recognition or a dedicated software platform for colaborative musical pieces. In any case we are open to listen your ideas.
If you want to colaborate on this project, you are invited to create a Pull Request with a descriptive text of the changes/updates.
You can also contact me:
[[email protected]](mailto:[email protected])
We hope you liked the project, if so, I invite you to leave a star ⭐, thanks for read :).