Models
A model defines the type of processing applied to your audio. Reference a model by its alias in a job target — and combine several models in a single job. Input lists the channel layouts a model accepts, and a tier can accept fewer than the model does, so check the note under a model before sending 5.1; see Audio formats for the 5.1 requirements.
GET /models — lists only the features enabled for your workspace, and a job that names a model outside them is refused with 403 FEATURE_NOT_ENABLED. Contact your Gaudio representative to enable a feature.Dialogue, Music & Effects (DME)
Split a mixed soundtrack into dialogue, music, and effects for dubbing, localization, and re-mixing. Request one or more of these tracks per job. DME models also accept 5.1 (six-channel) input and return 5.1 output.
dme_dialogue_v1DialogueDefault voice extraction that keeps all voices, including dialogue and vocals, while removing background music and effects. Typically delivers cleaner voice extraction than the dialogue-only model.
dme_dialogue_remove_vocals_v2DialogueDialogue-only extraction that removes vocals, music, and effects, using the second-generation dialogue model. It keeps the strengths of `dme_dialogue_remove_vocals_v1` and tends to do better in two situations — quiet speech such as walla, and dialogue in scenes with loud music or sound effects. Premium tier only.
dme_dialogue_remove_vocals_v1DialogueDialogue-only extraction that removes vocals, music, and effects. Use it when sung or music vocals overlap dialogue and you need the dialogue isolated from those vocals.
dme_music_v1MusicExtracts music, including vocals, while removing dialogue and effects.
dme_music_variety_v1MusicExtracts music, including vocals, while removing dialogue and effects, from variety and other unscripted entertainment — talk shows, reality, and studio programs where speech, laughter, and audience reaction run over the music. Tuned to keep that sound out of the music track. Standard tier only.
dme_effects_v2EffectsExtracts effects only, removing dialogue, music, and vocals, using the second-generation effects model. Premium tier only.
dme_effects_v1EffectsExtracts effects only, removing dialogue, music, and vocals.
dme_music_effects_v1MusicEffectsPreserves music, effects, and vocals while removing only dialogue. Together with `dme_dialogue_remove_vocals_v1` it reconstructs the original mix.
dme_music_effects_remove_vocals_v2MusicEffectsPreserves music and effects only, removing both dialogue and vocals, with cleaner separation and higher fidelity than `dme_music_effects_remove_vocals_v1`.
dme_music_effects_remove_vocals_v1MusicEffectsPreserves music and effects only, removing both dialogue and vocals. Typically delivers cleaner music-and-effects reconstruction than `dme_music_effects_v1`.
Stem Separation
Break a music track into individual music stems, including vocals and instruments. Each model produces one stem; request several model aliases in one job to receive multiple stems.
stem_vocals_v2Extracts the vocal stem using the second-generation vocal model. Use `karaoke_vocal_v1` instead when you want the karaoke-style vocal/accompaniment split.
stem_vocals_v1Extracts the vocal stem from the multi-stem separation family. Use `karaoke_vocal_v1` instead when you want the karaoke-style vocal/accompaniment split.
stem_drums_v2Extracts the drum and percussion stem using the second-generation drum model.
stem_drums_v1Extracts the drum and percussion stem from music.
stem_bass_v2Extracts the bass stem using the second-generation bass model.
stem_bass_v1Extracts the bass stem from music.
stem_guitar_v1Extracts the guitar stem from music.
stem_piano_v1Extracts the piano stem, acoustic and electric, from music.
Instrumental Track (Karaoke) Separation
Split a song into vocals and instrumental backing for karaoke, remixing, and vocal removal workflows. Each model produces one output stem.
karaoke_accompaniment_v1Extracts the instrumental backing track from music.
karaoke_vocal_v1Extracts the vocal track from the karaoke-style vocal/accompaniment split while removing the instrumental backing.
De-noise
Separate clean speech from background noise for dialogue cleanup, podcasts, and voice recordings. Each model produces one output stem.
denoise_speech_v1Isolates clean speech from background noise for dialogue cleanup, podcasts, and voice recordings.
denoise_noise_v1Extracts the removed background noise from speech recordings.
Text Sync
Align a written lyric to a recording and get timings back. Unlike the separation categories, these models take a SECOND input — the lyric, as a UTF-8 text asset referenced by `inputs.lyricsAssetId` — and return a JSON document rather than an audio track. Behaviour is selected with `params`; the values each model accepts are listed below.
lyrics_sync_v1Aligns a plain-text lyric to a song and returns per-line timings — and, at word sync level, per-word timings for karaoke highlighting.
inputs.lyricsAssetIdrequiredUTF-8 plain-text lyric, one line per lyric line.params.languagerequiredenkojaLanguage of the lyric.params.syncLevellineworddefault word`word` returns per-word timings; `line` returns line timings only, and its words carry no times.Accepts up to 30 minutes of audio — a longer input is rejected when the job is created.