PyFLAC - Real-time lossless audio compression in Python

Hello, I am trying to use pyflac on raspberry pi zero, but I am getting Illegal instruction error.

Collecting pyflac
  Downloading https://files.pythonhosted.org/packages/21/ec/621fd90f1b98477fa1c147edab0ea54cb8bd4b34773f698d5da665a197f3/pyFLAC-2.0.0.tar.gz (1.8MB)
     |████████████████████████████████| 1.8MB 1.0MB/s
Requirement already satisfied: cffi>=1.4.0 in /home/pi/.local/lib/python3.8/site-packages (from pyflac) (1.15.0)
Collecting SoundFile>=0.8.0 (from pyflac)
  Downloading https://files.pythonhosted.org/packages/eb/f2/3cbbbf3b96fb9fa91582c438b574cff3f45b29c772f94c400e2c99ef5db9/SoundFile-0.10.3.post1-py2.py3-none-any.whl
Requirement already satisfied: numpy in /home/pi/.local/lib/python3.8/site-packages (from pyflac) (1.22.4)
Requirement already satisfied: pycparser in /home/pi/.local/lib/python3.8/site-packages (from cffi>=1.4.0->pyflac) (2.21)
Building wheels for collected packages: pyflac
  Building wheel for pyflac (setup.py) ... done
  Created wheel for pyflac: filename=pyFLAC-2.0.0-cp38-cp38-linux_armv6l.whl size=1938080 sha256=30e340d066abeca1358fca4f26c1f82151e68ece9a1880f5345360b523eecd3b
  Stored in directory: /home/pi/.cache/pip/wheels/5a/96/19/863b328487fa805e47b53d7da91885d99770ceae1af630e38e
Successfully built pyflac
Installing collected packages: SoundFile, pyflac
Successfully installed SoundFile-0.10.3.post1 pyflac-2.0.0

Python 3.8.1 (default, May 25 2022, 20:18:49)
[GCC 8.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import soundfile
>>> import pyflac
Illegal instruction

Implements seek and tell callbacks so that the stream encoder can insert metadata at the start of the file.

Implements metadata callback, although currently the callback just receives a blob of cdata.

Makes streamable subset optional to allow the use of the full format settings.

Hi,

First of all thank you for the package and the examples!

I'm currently looking for methods to compress data from neural electrophysiology recorded from the brain. We use neural probes with many channels (e.g. 384) and I was wondering if there is a way to tweak the library to allow for more than 2 channels. Is it possible?

Thank you in advance! Alessio

• Added support for Linux arm64 architectures
• Added support for Darwin arm64 architectures (macOS Apple Silicon)
• Fixed Raspberry Pi Zero library (see #13)
• Updated FLAC library to v1.3.4

Changed logo url in README.rst to https://raw.githubusercontent.com/sonos/pyFLAC/develop/assets/logo-black.png Tested on MacOS Chrome, and MacOS + iOS Safari

I think this is in a good enough state now to review.

Would like to hear your opinions on the API, at the moment it's pretty much just copied exactly from the libFLAC design. You can see some examples of usage in

• examples/passthrough.py: illustrates both encoder/decoder callbacks by passing data from a WAV file through them both. It also verifies that the uncompressed output is exactly the same as the original WAV audio data.
• examples/stream.py: streams audio data from the microphone and passes it through the encoder, optionally saving the output to a FLAC file.
• pyflac/__main__.py: this is the main module for the package (which is also used for the CLI tools, as it makes things more accessible on Windows and if you don't install globally). It converts WAV -> FLAC files and vice versa.

Note: if you are changing things and want to develop locally, read CONTRIBUTING.rst. You have to explicitly build the CFFI .so (this is done automatically when using setuptools to install).

External contribution from @ali1234. Retested against previous benchmarks and it does not affect performance.

• Added seek and tell callbacks to StreamEncoder
• Renamed the write callbacks from callback to write_callback for StreamEncoder and StreamDecoder

https://github.com/sonos/pyFLAC/blob/19b3322851ac6e9e8abe2ddd4f285c90a52c7f97/tests/test_encoder.py#L127-L129

The third line here has no effect. If encoder._init() raises an exception (which we expect it to), all subsequent lines won't run at all.

The robot effect was due to the fact that the audio data passed to the sounddevice/portaudio callback, sometimes occupied the same memory region in consecutive callbacks. Which meant that by the time we processed it (after sleeping a little while), the audio blocks had been overwritten, leading to the stuttering kind of robot effect.

So I have reverted to a raw audio stream. Converting to bytes before adding to the queue guarantees that it has its own memory space so it won't get overwritten. Then we convert to a numpy array in the processing thread.

I chose this over doing a np.copy just because it's takes up less time in the high priority audio callback, which is not so much of an issue here. If you think it is more clear to do an np.copy here I will switch back 👍

• Ditch the decoder read_callback by buffering bytes passed in via the process method, and passing to the decoder's read callback internally. Making the encoder and decoder interfaces more alike.
• Rename write_callback to callback