Hi all,

 

Here's a small collection of tools to deal with Rocksmith songs:

https://github.com/x0l/rs-utils

 

Mostly intended for OSX users (command line batch convert), it can also be useful for others (make WEMs without launching Wwise).

 

I'm a bit lazy at the moment, but if I (or anyone else) find the time, I'll add JSON and aggregate graph generation.

 

Mostly python and shell code, easy to understand/port/improve.

 

Cheers,

Holy crap, CLI Wwise? Awesome, that program is a HUGE PITA to deal with, esp. since it doesn't redraw the screen controls (checkboxes/tabs/etc.) without moving the whole window.

Having some slight issues w/ the wav2wem command, or more specifically the WwiseCLI part. The program runs, converts the audio, but then never exits. So the mv & rm command never get executed, unless I manually kill the wwise/wine process. Also, the PostGen command (CopyStreamedFiles.exe) doesn't seem to work, so I just removed that from the .wproj & changed the mv line to grab the file from .cache/Windows/SFX/song_A6AAD7A8.wem - still, without the wine process exiting, that part won't get run. Any ideas? Here's what I get running the CLI command directly:

 

$ WINEDEBUG=-all wine "~/.wine/drive_c/Program Files/Audiokinetic/Wwise v2013.2.2 build 4828/Authoring/Win32/Release/bin/WwiseCLI.exe" Template.wproj -GenerateSoundBanks -ContinueOnError

 

Wwise | v2013.2.2 | Build no.4828 | © Audiokinetic Inc. 2006-2013. All rights reserved.

 

Application tried to create a window, but no driver could be loaded.

The explorer process failed to start.

 

*** Loading Project ***

 

Initializing...

Loading Project...

Loading Work Unit: Conversion SettingsDefault Work Unit.wwu

Loading Work Unit: Master-Mixer HierarchyDefault Work Unit.wwu

Loading Work Unit: Actor-Mixer HierarchyDefault Work Unit.wwu

Loading Work Unit: EventsDefault Work Unit.wwu

Loading Work Unit: SwitchesDefault Work Unit.wwu

Loading Work Unit: StatesDefault Work Unit.wwu

Loading Work Unit: SoundBanksDefault Work Unit.wwu

Loading Work Unit: Game ParametersDefault Work Unit.wwu

Loading Work Unit: EffectsDefault Work Unit.wwu

Loading Work Unit: PresetsDefault Work Unit.wwu

Loading Work Unit: Soundcaster SessionsDefault Work Unit.wwu

Loading Work Unit: QueriesDefault Work Unit.wwu

Loading Work Unit: Interactive Music HierarchyDefault Work Unit.wwu

Loading Work Unit: TriggersDefault Work Unit.wwu

Loading Work Unit: AttenuationsDefault Work Unit.wwu

Loading Work Unit: Dynamic DialogueDefault Work Unit.wwu

Loading Work Unit: Mixing SessionsDefault Work Unit.wwu

Updating Project...

Finalizing...

 

*** Bank Generation ***

 

[Collecting SoundBank Information...] RS_SOUNDBANK

Message 0       Evaluation mode: SoundBanks contain 1 media item(s) out of the 200 permitted.   Windows«

[Collecting SoundBank Information...] Init

[Collecting SoundBank Information...] RS_SOUNDBANK

[Converting files...] RS_SOUNDBANK

[song.wav] Applying conversion settings...

[song.wav] Encoding Vorbis...

[Converting files...] Init

[Converting external sources...] Reading external sources file

[Converting external sources...]

[Generating SoundBank Media Sections...] RS_SOUNDBANK

[Writing to disk...] Z:Users[...]TemplateGeneratedSoundBanksWindowsRS_SOUNDBANK.bnk

[Generating SoundBank Hierarchy Sections...] RS_SOUNDBANK

[soundBank Log Info] Name: RS_SOUNDBANK Platform: Windows« | Language: SFX  | Created:Yes | Status: Successful

 

Process completed successfully.

 

...and then it just hangs there until the process is killed. -_-

Hey! Let's create a GUI?!

Tkinter or Qt?

 

--

 

It works only on MAC?

@@th1rtyf0ur

You should install dotnet35sp1 to get-post processing working

 

I've no idea why it hangs up though.

@@SrMouraSilva

It's mostly cross-platform since it's python. Shell script need to be translated in .bat files for Win

@@th1rtyf0urYou should install dotnet35sp1 to get-post processing working I've no idea why it hangs up though.

I think it hangs because of the wrong Wwise version, 2013.1.3 failed spectacularly, 2013.2.2 probably thought is compatible. Here are the links:https://www.audiokinetic.com/files/?get=2013.2.4_4836/Authoring_Win32.msihttps://www.audiokinetic.com/files/?get=2013.2.4_4836/Authoring_Data.msiI have only dotnet40 for toolkit. A dialog babbling something about unknown runtime pops up (missing dotnet35 apparently) and then it ends with "Process completed with warning(s)." message but WEM is there. 

[SoundBank Log Info] Name: RS_SOUNDBANK Platform: Windows« | Language: SFX  | Created:Yes | Status: Successful[Running Post-generation step: Copy Streamed Files...] Executing command: "C:Program FilesAudiokineticWwise v2013.2.4 build 4836AuthoringWin32ReleasebintoolsAkCopyStreamedFiles.exe" -info "Z:homexTemplateGeneratedSoundBanksWindowsSoundbanksInfo.xml" -outputpath "Z:homexTemplateGeneratedSoundBanksWindows" -banks "RS_SOUNDBANK Init" -languages "English(US)"Warning	29	Custom pre-generation or post-generation executable returned a non-zero value: '"C:Program FilesAudiokineticWwise v2013.2.4 build 4836AuthoringWin32ReleasebintoolsAkCopyStreamedFiles.exe" -info "Z:homexTemplateGeneratedSoundBanksWindowsSoundbanksInfo.xml" -outputpath "Z:homexTemplateGeneratedSoundBanksWindows" -banks "RS_SOUNDBANK Init" -languages "English(US)"', error id: '-2146232576'	Windows«Process completed with warning(s).

@@shrubbery

 

Thanks for pointing this out, I only used 2013.2.4

 

I think you need .Net 3.5 sp1 in oder to have post-processing work and bnk generation. My initial plan was to get both the WEM and the BNK from WwiseCLI (and the _preview files too). It's (mostly) a matter of hard coding the right ID and ShortID in the template. The only real issue is midi events, I haven't figure out how to configure this in Wwise

I don't really care about replacing toolkit code now, I would like to replace EoF by TuxGuitar entirely and with all RS features... but I don't see myself diving into Java anytime soon

Woo! If you guys need a mac development forum PM me.

@@shrubbery

 

:) I was planing something similar using Guitar Pro and Go Play Along for sync.

 

My ultimate goal would be being able to build a PSARC from:

- GP tab

- Sync points from GPA

- a few metadata

 

I don't care for DD, or lyrics or showlights...

 

The code already exists in EoF, it's a (painful) matter of extracting it

That's actually what I have in mind. It would need to be an extension to tab format (probably TG native) to hold extra RS information (techniques, per track phrases and sections, tone switching) and GPA sync would be only applied on export.

 

Basically reimplementation of tab to RS XML conversion is the huge part of what would be necessary to have it working. I don't care about manually created DD either but it wouldn't be impossible.

 

BTW I know how to extract sync points from GPA database.

GPA sync points Measure + fraction of beats. Here's quick copy/paste

import structfor e in z:    s = e[9]    if s is None:        continue    inc = 0    while s[inc] != 'x00':        inc += 1    n = struct.unpack('>I', s[inc:inc+4])[0]    inc += 4    for i in range(n):        time             = struct.unpack('>f', s[inc   :inc+4 ])[0] / 1000.0        measure          = struct.unpack('>I', s[inc+4 :inc+8 ])[0]        measure_fraction = struct.unpack('>f', s[inc+8 :inc+12])[0]        beat_duration    = struct.unpack('>f', s[inc+12:inc+16])[0] / 1000.0        inc += 16        print time, measure, measure_fraction, beat_duration 

EDIT: Read your post too quickly :)

Yeah, I said I know, I had to write the code to shift it :)

showlights...

 

Its handled by toolkit now so forget about it :P

[update] graph, hsan, json generation

you just done cli version of toolkit for mac...

Yup :)

Portable (almost), and much much fewer LoC...

 

It's not quite useable right now, it soon will be.

 

Next big step will be to convert from GP to RS XML in one click (or rather one command) ie rewrite the part of EoF I'm really interested in ;) Ruby is our friend since it already provides a nice Guitar Pro gem.

Well then I'll try to include this gp>RS2 xml tools to rs toolkit, cause its so tasty :3

@@xavatar so you won't consider using TuxGuitar framework? It already knows how to read GP too (and does it very well) and functionality can be further enhanced beyond what will ever be possible in unmodified GP and EOF import.

Mostly python and shell code, easy to understand/port/improve.

 

Thanks for this. Helps me negate a tedious step of the process.

@@shrubbery

 

Nope :)

I don't want to deal with java for the same reasons I don't like mono. At the end of the day, all I'm interested in is to read a GP tab.

 

Aren't those possibilities also beyond Rocksmith too ?

Of course not, just as one example, with plain GP file you will need to handle phrases and sections with awkward workarounds (possibly with multiple copies of the file like I have to with EOF now).

I see you're still missing tone support? I was going to suggest extraction from profile like @@Alex360 did, really useful, but there is no SNG compilation support for tones at all?

Cause tones are read from jsom files an they got no descriptors in profile, need to set $[-1] but toolkit set it yo bass as default :(

tones sng compilation should be ok

missing json part though. Actually i believe tone definitions belong to the xml.

 

 

I just added a profile and psarc -> json tool for tones

tones.py /path/to/dlc/*.psarc /path/to/profile/*_prfldb > tones.json