irmen / pyminiaudio

Hi everyone,

I am stuck with an issue, I have many audio files which are in mp4 format with the codec MPEG-4 HE AAC. They are quite a long files so I use the function stream_file(...) to read chunks but I get the error ('failed to init decoder', -1).

A little investigation shows that it comes from this section of the code

pyminiaudio/miniaudio.py

Line 1022 in 55614d3

if result != lib.MA_SUCCESS:

My understanding is that lib.ma_decoder_init_file(...) is not able to decode the file. I am not very familiar with audio files codecs so I am not even sure if what I am requesting actually makes sense.

All I know is that I am just too lazy to convert them to mp3 with ffmpeg and that they seem to take less disk space in their current format

I see in this tutorial that there is a way around to use ffmpeg to read a format that is not natively supported.

import subprocess
import miniaudio

channels = 2
sample_rate = 44100
sample_width = 2  # 16 bit pcm
filename = "samples/music.m4a"  # AAC encoded audio file

def stream_pcm(source):
    required_frames = yield b""  # generator initialization
    while True:
        required_bytes = required_frames * channels * sample_width
        sample_data = source.read(required_bytes)
        if not sample_data:
            break
        print(".", end="", flush=True)
        required_frames = yield sample_data

with miniaudio.PlaybackDevice(output_format=miniaudio.SampleFormat.SIGNED16,
                              nchannels=channels, sample_rate=sample_rate) as device:
    ffmpeg = subprocess.Popen(["ffmpeg", "-v", "fatal", "-hide_banner", "-nostdin",
                               "-i", filename, "-f", "s16le", "-acodec", "pcm_s16le",
                               "-ac", str(channels), "-ar", str(sample_rate), "-"],
                              stdin=None, stdout=subprocess.PIPE)
    stream = stream_pcm(ffmpeg.stdout)
    next(stream)  # start the generator
    device.start(stream)
    input("Audio file playing in the background. Enter to stop playback: ")
    ffmpeg.terminate()

I imagine that this strategy could be used. In my case I want 15s long chunks but I am not sure how to make that to work...
I get an error while trying to iterate over the stream object

    required_bytes = required_frames * channels * sample_width
TypeError: unsupported operand type(s) for *: 'NoneType' and 'int'

In general: builtin AAC decoding is never going to be available unless it will be in the miniaudio C library, which is highly unlikely. So yes as you discovered, use an external decoder to use this audio file format.

can you post the entire stack trace of that last error? I don't know where it's coming from. In any case, one of those three variables is unknown somehow.
"i want 15 seconds chunks" this sounds like a file conversion issue rather than playback. While miniaudio contains some rudimentary support for conversion of audio files, I suggest to simply use ffmpeg only to do this task

closing this now.

add the possibility to read aac encoded audio file