So yeah, I should be working on Rizelmine or some other stuff, but I've been trying to solve a vexing puzzle for a future project. As I mentioned in the past, I want to re-release the Summer 2002 series that Master of Entertainment sandwiched between the two halves of Rizelmine, G-On Riders. I managed to obtain
these DVD raws, which are a definite improvement over the TV-rips, but are unfortunately encoded in .wmv with WMV3 video and Microsoft
ACM audio. After changing the file names to remove Japanese characters, I managed to feed ep 01 into GraphEdit, insert the Haali Matroskamuxer filter, and create an .mkv for future muxing.
The obstacle here is the audio -- it won't load in Aegisub for retiming, it can't be extracted "as-is" by mkvextract, and it refuses to be transcoded. I was able to isolate the audio by remuxing the .mkv in mkvmerge without the video, although that raises questions of what the correct file extension should be. (.mka? .wav? .wma? .acm?)
The resulting audio file can be found
here.So, viewers, leechers, and citizens of the broader Internet, I turn to you for help: can anyone find a way to transcode this POS ancient audio track into something usable, like .mp3, ogg, or aac? As far as settings go, 192 kbps should be more than enough.
Programs I've attempted and failed with:
Audacity
Free Audio Converter
Avidemux
AnalogX Converter
AoA Audio Extractor Basic
Handbrake (no surprise that it failed)
Many of them will allow the file to be loaded, but pressing convert brings up an instantly-complete progress window, and results in a 0-byte or very small ( <1 MB) audio file. Free Audio converter actually took some time in "Processing," but then seems to hang forever after the progress bar completes.
If the audio can't be transcoded, I see my options as:
1) Use the 128 kbps mp3 audio from the TV-rips, and hope/pray that everything lines up properly. Lots of risk of desync, plus it's low-bitrate audio encoded back in 2002.
2) Play the audio file itself or the original raw in a media player, use Audacity to record the audio to another format in real-time, and remux with the video. Not only would that be time-consuming, but I'd have to compensate for human-error delays between me starting the file and starting the recording.