Skip to main content

Model Overview

Stem Separation

ModelAPI ReferenceDescription
gsep_music_hq_v1ViewSeparates vocals, drums, bass, electric guitar, piano, and other instruments
gsep_music_shq_v1ViewHigher-quality model that separates into vocals and accompaniment only
gsep_speech_hq_v1ViewIsolates speech from background noise to deliver clearer voice

DME Separation

ModelAPI ReferenceDescription
gsep_dme_dtrack_v1ViewOutputs only the dialogue track, removing both music and effects. Ideal for subtitling, dubbing, and localization
gsep_dme_etrack_v1ViewOutputs only the effects track, removing dialogue and music. Helpful for Foley, sound design, and post-production editing
gsep_dme_metrack_v1ViewOutputs music + effects together, with dialogue removed. Perfect for ADR (Automated Dialogue Replacement)
gsep_dme_mtrack_v1ViewOutputs only the music track, without dialogue or effects. Great for karaoke, remixing, or music replacement

AI Text Sync

ModelAPI ReferenceDescription
gts_lyrics_line_v1ViewLine-level lyric alignment; supports English, Korean, Japanese, Chinese (Simplified)
gts_lyrics_line_v3Not yet available