Skyrim Audio Module

Handle Skyrim audio file operations including FUZ, XWM, and WAV formats using Spooky's AutoMod Toolkit.

Prerequisites

Run all commands from the toolkit directory:

bash

cd "<TOOLKIT_PATH>"
# Example: cd "C:\Tools\spookys-automod-toolkit"

Overview

Skyrim uses several audio formats:

Format	Extension	Purpose
FUZ	.fuz	Combined voice audio + lip sync
XWM	.xwm	Compressed audio (xWMA codec)
WAV	.wav	Uncompressed audio
LIP	.lip	Lip sync data

Voice files are typically stored as FUZ (XWM audio + LIP lip sync combined).

Command Reference

Get Audio Info

bash

dotnet run --project src/SpookysAutomod.Cli -- audio info "<audio>"

Input	Description
`<audio>`	Path to .fuz, .xwm, or .wav file

Output includes: File type, file size, audio data size, lip sync presence (for FUZ).

Extract FUZ

bash

dotnet run --project src/SpookysAutomod.Cli -- audio extract-fuz "<fuz>" --output "<dir>"

Extracts FUZ into XWM and LIP components.

Option	Description
`--output`, `-o`	Output directory

Output files:

•filename.xwm - Audio data
•filename.lip - Lip sync data (if present)

Create FUZ

bash

dotnet run --project src/SpookysAutomod.Cli -- audio create-fuz "<xwm>" --output "<file>" [options]

Option	Description
`<xwm>`	Path to XWM audio file
`--output`, `-o`	Output FUZ file path
`--lip`	Path to LIP file (optional)

Convert WAV to XWM

bash

dotnet run --project src/SpookysAutomod.Cli -- audio wav-to-xwm "<wav>" --output "<file>"

Option	Description
`<wav>`	Path to WAV file
`--output`, `-o`	Output XWM file path

WAV Requirements:

•PCM format
•16-bit
•Mono or stereo
•44100 Hz recommended for voice

Common Workflows

Inspect Voice File

bash

# Check FUZ file info
dotnet run --project src/SpookysAutomod.Cli -- audio info "./Sound/Voice/MyMod.esp/NPC/Line01.fuz"

# Check if it has lip sync data
# Output will show hasLipSync: true/false

Extract Existing Voice Files for Analysis

bash

# 1. Extract from BSA if needed
dotnet run --project src/SpookysAutomod.Cli -- archive extract "SomeMod.bsa" --output "./Extracted" --filter "sound/*"

# 2. Extract FUZ to components
dotnet run --project src/SpookysAutomod.Cli -- audio extract-fuz "./Extracted/Sound/Voice/SomeMod.esp/NPC/Line01.fuz" --output "./AudioDebug"

# Creates:
#   ./AudioDebug/Line01.xwm
#   ./AudioDebug/Line01.lip (if present)

Create Voice Line (Full Workflow)

bash

# 1. Record/obtain WAV file (44100 Hz, 16-bit, mono recommended)

# 2. Convert WAV to XWM
dotnet run --project src/SpookysAutomod.Cli -- audio wav-to-xwm "./Source/Line01.wav" --output "./Audio/Line01.xwm"

# 3. Generate LIP file (requires FaceFX or external tool - not included)
#    Or use existing LIP from similar dialogue

# 4. Create FUZ with lip sync
dotnet run --project src/SpookysAutomod.Cli -- audio create-fuz "./Audio/Line01.xwm" --lip "./Audio/Line01.lip" --output "./Sound/Voice/MyMod.esp/NPC/Line01.fuz"

# Or without lip sync (for sound effects)
dotnet run --project src/SpookysAutomod.Cli -- audio create-fuz "./Audio/Line01.xwm" --output "./Sound/Voice/MyMod.esp/NPC/Line01.fuz"

Create Sound Effect (No Lip Sync)

bash

# 1. Convert WAV to XWM
dotnet run --project src/SpookysAutomod.Cli -- audio wav-to-xwm "./Source/effect.wav" --output "./Sound/FX/MyMod/effect.xwm"

# Or package as FUZ without LIP (some mods prefer this)
dotnet run --project src/SpookysAutomod.Cli -- audio wav-to-xwm "./Source/effect.wav" --output "./temp/effect.xwm"
dotnet run --project src/SpookysAutomod.Cli -- audio create-fuz "./temp/effect.xwm" --output "./Sound/FX/MyMod/effect.fuz"

Batch Extract Voice Files

bash

# Extract all FUZ files from a mod's BSA
dotnet run --project src/SpookysAutomod.Cli -- archive extract "VoiceMod.bsa" --output "./Extracted" --filter "*.fuz"

# Then extract each FUZ individually to analyze
# (Loop through files in script)

Voice File Organization

Skyrim expects voice files in specific locations:

code

Data/
  Sound/
    Voice/
      MyMod.esp/              # Must match plugin name exactly
        NPC_EditorID/         # Must match NPC's Editor ID
          DialogueTopic_ResponseNumber.fuz

Example path:

code

Data/Sound/Voice/MyMod.esp/MerchantBob/DialogueMerchantGreeting_00001234_1.fuz

Voice File Naming Convention

code

<TopicEditorID>_<FormID>_<ResponseNumber>.fuz

•TopicEditorID - The dialogue topic's Editor ID
•FormID - 8-digit hex form ID of the dialogue response
•ResponseNumber - Response index (usually 1)

FUZ File Format

FUZ files combine audio and lip sync:

code

Header:
  Magic: "FUZE" (4 bytes)
  Version: uint32
  LIP size: uint32

Data:
  LIP data: [LIP size bytes]
  XWM data: [remaining bytes]

Audio Recommendations

For Voice Lines

•Sample Rate: 44100 Hz
•Bit Depth: 16-bit
•Channels: Mono (recommended) or Stereo
•Format: PCM WAV before conversion

For Sound Effects

•Sample Rate: 44100 Hz or 22050 Hz
•Bit Depth: 16-bit
•Channels: Mono or Stereo depending on effect
•Format: PCM WAV before conversion

For Music

•Sample Rate: 44100 Hz
•Bit Depth: 16-bit
•Channels: Stereo
•Format: Use XWM or vanilla formats

Limitations

This module CAN:

•Read FUZ/XWM/WAV file information
•Extract FUZ to components
•Create FUZ from components
•Convert WAV to XWM

This module CANNOT:

•Generate LIP (lip sync) files
•Convert XWM back to WAV
•Edit audio content
•Generate voice from text

For LIP generation, use:

•FaceFX (included with Creation Kit)
•LipGen community tools
•xVASynth for AI voice generation (separate tool)

Important Notes

•LIP files required for good lip sync - Without LIP, NPCs won't move lips
•Path must match plugin name - Sound/Voice/PluginName.esp/
•NPC folder must match Editor ID - Case sensitive
•FUZ preferred for voice - Combines audio and lip sync cleanly
•Use --json flag for machine-readable output when scripting

JSON Output

All commands support --json for structured output:

bash

dotnet run --project src/SpookysAutomod.Cli -- audio info "./Sound/Voice/MyMod/Line01.fuz" --json

Example response:

json

{
  "success": true,
  "result": {
    "fileName": "Line01.fuz",
    "type": "FUZ",
    "fileSize": 45678,
    "audioSize": 42000,
    "hasLipSync": true,
    "lipSyncSize": 3678,
    "sampleRate": 44100,
    "channels": 1,
    "bitsPerSample": 16
  }
}