Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/kurtamohler/freqbench

Frequency response test bench for audio electronics
https://github.com/kurtamohler/freqbench

Last synced: 7 days ago
JSON representation

Frequency response test bench for audio electronics

Awesome Lists containing this project

README 

        
# FreqBench



Frequency response test bench for audio electronics



If you want to measure a frequency response curve for some audio signal

processing device (like a guitar pedal), this tool can do that for you.



## Hardware Setup



Obviously, you must have an audio electronic device that you want to test.

We'll refer to this device as the DUT (device under test).



Next, you'll need a real time audio interface. Something like a Focusrite

Scarlett 2i2 will work. If you don't want to buy a separate audio interface,

you could probably get away with using your computer's builtin audio interface.

However, (at least on most Linux distros) you would probably need to install

and run a real-time audio driver like [JACK Audio Connection

Kit](https://jackaudio.org/).



Plug the output of your audio interface into the input of the DUT. Then plug

the output of the DUT to the input of the audio interface. This will allow us

to generate a signal with your computer, send it through the DUT, capture the

DUT's output, and process it with your computer.



## Software Requirements



[Conda](https://docs.conda.io/projects/conda/en/latest/user-guide/install/index.html) must be installed.



## Installation



Run the following command to create and activate a conda environment.



```

$ conda env create -n freqbench -f environment.yaml

$ conda activate freqbench

```



Install the freqbench module.



```

$ python setup.py install

```



## Usage



First, you must find out the input and output device IDs for your audio

interface. You can run the following in your terminal.



```

$ python

>>> import freqbench

>>> freqbench.get_devices()

```



This gives us all the audio input and output devices available on the system.

The number assocated with each of them on the left hand side is the ID of the

device.



Here is an example output:



```

DevicesInfo(

  input:

    6: "Scarlett 2i2 USB: Audio (hw:2,0)"

    7: "sysdefault"

    12: "default"

  output:

    0: "HDA NVidia: HDMI 0 (hw:1,3)"

    1: "HDA NVidia: HDMI 1 (hw:1,7)"

    2: "HDA NVidia: HDMI 2 (hw:1,8)"

    3: "HDA NVidia: HDMI 3 (hw:1,9)"

    4: "HDA NVidia: HDMI 4 (hw:1,10)"

    5: "HDA NVidia: HDMI 5 (hw:1,11)"

    6: "Scarlett 2i2 USB: Audio (hw:2,0)"

    7: "sysdefault"

    8: "front"

    9: "surround40"

    10: "surround51"

    11: "surround71"

    12: "default"

    13: "dmix"

)

```



In this example, I will use the Scarlett 2i2 interface. We have to take

note that the input device ID is 6 and the output device ID is 6. Note that

the input and output IDs may not always be equal to each other. It depends on

your setup.



Now we can go ahead and generate a test signal:



```python

>>> freq0 = 0

>>> freq1 = 22_000

>>> time = 10

>>> frame_rate = 44_100

>>> input_signal = 0.1 * freqbench.signal.sweep(freq0, freq1, time, frame_rate)

```



This generates an audio signal that sweeps through 0 Hz to 22 kHz over 10

seconds, using a frame rate of 44.1 kHz. Notice that we've multiplied the

signal by 0.1. `freqbench.signal.sweep` generates a signal at maximum volume,

which is bound to cause some clipping, so it's important to decrease the

volume by scaling it down.



If you're curious about what this sounds like, you can play it back and

listen. You can either play it through a different device on your system or

you can unplug the output of your audio interface from the DUT and attach it

to a speaker.



On my system, my computer's speakers are output device 12, "default" from the

`get_devices` printout above, so I will use that device. Note that if you

do something similar, and your system's audio driver is not real-time, there

may be some odd audio artifacts that vary each time you play it, like popping

and short static bursts. Or with some audio devices that are not real-time, it

may not even seem like it's playing at all. These artifacts aren't actually

part of the audio signals. If you play them through a real-time interface,

you'll notice a lot better consistency.



```python

>>> freqbench.play_signal(input_signal, 12, 44_100)

```