How to concat multiple streamed audio inputs

[jaredculp]

Thank you for this wonderful library! I am wondering how I can translate the code below to be able to concatenate multiple opus audio files using the BytesIO API.

In this contrived example I am reading files off disk, but in reality I am reading files from S3. I am trying to avoid any intermediate writes to disk and instead just stream directly to ffmpeg. Is this possible?

from pathlib import Path

import ffmpeg

input = Path.cwd().joinpath("input.txt")
with input.open("w") as f:
    # in reality, this is more like:
    # for i, obj in enumerate([
    #     s3.get_object(Bucket=bucket_name, Key=obj["Key"])["Body"].read()
    #     for obj in s3_bucket_list_objects_v2(bucket_name=bucket_name, key=key)["Contents"]
    # ]):
    for i, obj in enumerate(Path.cwd().glob("recording*.webm")):
        f.write(f"file '{content}'\n")

output = Path.cwd().joinpath("combined.webm")
ffmpeg.input(str(input), format="concat", safe=0).output(str(output), c="copy").run()