opus | libopus for VS2015 and ffmpeg build | Audio Utils library

There is not really a need for a subroutine. Delayed variable expansion is needed finally, but it is possible to first assign the fully qualified file name to an environment variable like FileName to avoid troubles with file names containing an exclamation mark.

The rewritten code according to the code posted in the question with some comments:

It might be missing

This happens if you are trying to open a file, but your path is a folder.

This can happen easily by mistake.

To defend against that, use:

how can I turn this array into a proper .opus or .ogg file?

You're capturing a Webm file stream via audio/webm;codecs=opus, so that would need to be converted from Webm to Ogg Opus. Instead, you could simply download or send the Webm Opus stream as a file as-is:

MDN Says:

For example, for any MIME type whose main type is text, you can add the optional charset parameter to specify the character set used for the characters in the data. If no charset is specified, the default is ASCII (US-ASCII) unless overridden by the user agent's settings. To specify a UTF-8 text file, the MIME type text/plain;charset=UTF-8 is used.

So, for anything based on text/... you can optionally add the charset.

https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/MIME_types#structure_of_a_mime_type

The following update to contentType() function demonstrates one solution.

A for /F command with a set to process the output of a command line results in starting in background one more Windows command processor with %SystemRoot%\System32\cmd.exe /c and the command line appended as additional arguments. Therefore delayed variable expansion is not enabled for this command process running in background as it would be necessary for echo !mkvmergeinfo!. It would be necessary to run the background cmd.exe with option /V:ON left to option /c and the command line to execute. But that is not possible, except running two additional cmd.exe, the first one with /c and a command line which starts with %ComSpec% /D /V:ON /C ... a second cmd.exe as suggested by aschipfl in his comment above.

I suggest to run mkvmerge.exe -i once for each *.mkv file and process the output using just internal commands of cmd.exe like FOR and IF without using find.exe and findstr.exe at all.

Example:

As you can see in strace, the execve command executes as: execve("/bin//nc", ["/bin//nc", "/bin//nc-e //bin/bash -nvlp 4455"], NULL) = 0 It seems to be taking the whole /bin//nc-e //bin/bash -nvlp 4455 as a single argument and thus thinks it's a hostname. In order to get around that, the three argv[] needed for execve() is pushed seperately. argv[]=["/bin/nc", "-e/bin/bash", "-nvlp4455"] These arguments are each pushed into edx, ecx, and ebx. since ebx needs to be /bin/nc, which was already done in the original code. we just needed to push 2nd and 3rd argv[] into ecx and edx and push it into stack. After that we just copy the whole stack into ecx, and then xor edx,edx to set edx as NULL.

Here is the correct solution:

The 8k video you are trying to download uses AV1 https://en.wikipedia.org/wiki/AV1 https://aomediacodec.github.io/av1-isobmff/#codecsparam coding format

To better understand it, video codec av01.0.16M.08 means the following

• av01 is the AV1 codec
• 0 indicates Profile 0
• 16M indicates level 6.0 identifier 16
• 08 indicates an 8 bit bit depth

The key is in the level identifier 16 (6.0) A level 4.0 decoder can decode 1080p video @30fps but it won't be able to decode 1080p@60fps and it requires level 4.1 Similarly for 4k@30fps we need level 5.0 and 60fps need level 5.1 and 8k requires a level 6.0 or higher

Refer below for more information about these levels

https://en.wikipedia.org/wiki/AV1#Levels

Level 7 has not been defined yet.

You can see that some of these levels are fairly new as 8k is yet to be mainstream. ffmpeg only started supporting AV1 codec from version 4.0 and more decoder supports are being added as and when they come in newer versions.

So, in order to answer your question, you just need to update your ffmpeg to the latest version and it should solve the issue. Or you can download a video with a lower decoder level requirement and it will suffice. The added information is for reference so that someone who stumbles upon the question can better understand why the issue has happened and I hope it will help troubleshoot similar issues in the future.

Marian is an open-source tool for training and serving neural machine translation, mostly developed at the University of Edinburgh, Adam Mickiewicz University in Poznań and at Microsoft. It is implemented in C++ and is heavily optimized for MT, unlike PyTorch-based Huggingface Transformers that aim for generality rather than efficiency in a specific use case.

The NLP group at the University of Helsinki trained many translation models using Marian on parallel data collected at Opus, and open-sourced those models. Later, they also did a conversion of the trained model into Huggingface Transformers and made them available via the Huggingface Hub.

MarianMT is a class in Huggingface Transformers for imported Marian models. You can train a model in Marian and convert it yourself. OpusMT models are Marian models trained on the Opus data in Helsinki converted to the PyTorch models. If you search the Huggingface Hub for Marian, you will find other MarianMT models than those from Helsinki.