External Tools
Yass Reloaded integrates with several external tools to extend song creation and editing workflows.
Core Tools
FFmpeg
Used for:
- audio conversion
- broad file format support
- preprocessing for analysis and transcription
WhisperX
Used for:
- local transcription
- timestamped word alignment
- note alignment support
- isolated Python runtime when needed
OpenAI
Used for:
- transcript-based note alignment workflows
LrcLib
Used for:
- online lyric lookup in the song creation wizard
- importing ready-made lyric text and timed lyric lines when available
- providing a lightweight fallback before full transcription is needed
MVSEP
Used for:
- cloud-based stem separation
- API-key based online vocal/instrumental separation
- choosing a Vocal Model and Audio Format for returned stems
audio-separator
Used for:
- local stem separation
- optional model caching
- isolated Python runtime when needed
fanart.tv
Used for:
- cover lookup workflows from the library
- fetching alternate cover candidates for a song folder
USDB
Used for:
- direct song search against the live USDB service
- compare and review workflows against the current USDB version of a song
- importing songs from USDB into the local library
- pushing reviewed edits back to USDB when the current account has edit rights
USDB Syncer
Used for:
- local USDB song search via the Syncer SQLite database
- refreshing the locally cached USDB song list
- supporting
.usdb-based links between local songs and USDB entries
Configuration
Most tool setup is available under:
Settings -> External Tools
Python-based tools use a two-level runtime model:
Settings -> External Tools -> Locationscan define a sharedDefault Python executable- WhisperX and audio-separator can each use their own tool-specific Python runtime
- if a tool-specific Python field is empty, Yass falls back to the shared default Python
- if
Update packageis used while the tool-specific Python field is still empty, Yass creates a dedicated virtual environment for that tool and stores its Python path automatically
The fanart.tv API key is configured in:
Settings -> Library -> Setup
USDB account-related setup is configured in:
Settings -> Library -> Setup
LrcLib does not require a dedicated API key or a separate setup page.
The wizard defaults for choosing local vs online engines are configured in:
Settings -> Wizard
The USDB Syncer path is configured in:
Settings -> External Tools -> Locations
Health Checks
Several tool integrations provide health checks to verify:
- executable paths
- package availability
- FFmpeg presence
- WhisperX runtime suitability
- audio-separator package availability and script resolution
Recommended Mindset
Not every workflow needs every tool:
- use FFmpeg as the baseline
- add WhisperX for local transcription and timing support
- use OpenAI for cloud transcription workflows
- use LrcLib first when you mainly want existing lyrics with minimal setup
- use MVSEP or audio-separator when you need stems
- use direct USDB integration when you want compare/import/edit workflows against the live song database
- use fanart.tv for library polish and media management
- use USDB Syncer when you want faster USDB search and a local DB-backed lookup path
Goal of This Section
This section will gradually expand into dedicated setup guides for each supported integration.