Shenmue Re-dubbing Project

[Francko1976]

Hello everyone,
This is my first time posting here. A long time ago, I used to translate Winning Eleven from Japanese using the beloved "patches."

I’m looking for help from anyone who has done this or has tried.

I’ve read and learned a lot from the "Shenmuesubs" and "Shenmuedojo" websites, but I couldn’t find anyone explaining how to re-encode the STR audio files.

Extracting them was easy with the "ADPCM Streaming Toolkit".
The files are processed and converted into mono 22Hz WAV files.

I replaced the track, keeping the duration unchanged, with a new audio track.
Then, I tried several "techniques"...
STR contains a SPSD header file. So I suppose the audio is encoded in that format.
The first, and perhaps the most correct, was using a tool called "WAV2SPSD" by VincentNL, which was designed for Sega Naomi ROMs. It requires a "stereo" WAV audio file, which it then converts back into "STR" in two different modes for Naomi.
WAV2SPSD uses two executables I downloaded from the Katana Dreamcast SDK GitHub: waveconv.exe and MkStream.exe.

Next, I copied the first 20 bytes from the original STR file to the new STR. I read this around the web.
I also tested it with ADPCM Streaming to see if it would extract it back to WAV, and it did.

I reinserted the STR file into the AFS archive, recreated the index, and tested the game. Unfortunately, the audio remains silent.

Does anyone know how to help me?

My intention is to redo the entire audio dubbing for Shenmue with AI, cloning the original Japanese voices, or maybe other languages

thanks in advance people!

[Francko1976]

I did it
the problem was the idx rebuld... and not the re-encoded str.

https://youtu.be/2qRFDnZbYIA