Studio
Browse API endpoints
post/presets/run

Run any Neural Analog audio preset

Choose an operation and a preset from GET /presets. The same request shape can clean an AI song, master a mix, isolate sounds, or turn audio into MIDI.

Get an API keyView examples

How this endpoint works

Use this as the default processing endpoint for new integrations and AI agents. Set operation to restoration, mastering, stem_split, or transcription. To chain jobs, set source_type and source_id so the next preset processes the completed result instead of the original audio.

  1. 1

    Choose from the catalog

    View presets in GET /presets, filter by the requested operation, and choose its default or best-matching use case.

  2. 2

    Start one processing job

    Send the audio ID, operation, preset, source selection, and only the parameters advertised for that preset.

  3. 3

    Poll and deliver the result

    Use object_type and object_id from the response to poll status, then download the completed artifact.

Common use cases

Choose this operation when it matches the source and result your pipeline needs.

Clean up an AI-generated song

Import a Suno or Mureka link, run the default restoration preset, and return the completed WAV.

Process a derived result

Master a restoration, repair one isolated stem, or transcribe a separated instrument by selecting that source object.

Expose one agent action

Give an AI one processing operation instead of separate restoration, mastering, stem, and MIDI actions.

Processing used by 80,000+ music makers

Results people rely on

“I've been loving the app. The audio restoration works amazing on my suno songs”

Jordi

Audio Super Resolution

“Amazing tool for audio. Clean, simple, and effective. I would spend hours in RX to get the same results. Give it a try. I can save you hours of production time.”

Dan Campbell ~Riffster

“It can bring back clarity without making everything sound too artificial or overprocessed. Stem separation is also surprisingly clean.”
R

R.

Audio restoration and stem separation

“Does amazing job in restoring Audio from bad quality MP3s and badly extracted audio from CDs. Always my go to for audio restoration.”
N

N.

MP3 and CD audio restoration

Restoration · Choose what you want to fix

MP3 Music Restoration (Apollo, 2025)

preset="universal_enhancer"

This model was trained to regenerate high-quality music from its low-quality MP3 version. It fills gaps in the spectrogram and restores missing frequencies above 16 kHz. Perfect for your old 128 kbps MP3s downloaded from the internet and lost media.

Parameters
stereo_modeStereo Mode

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.

AI Cover - Stable Audio 3 (Stable Audio 3 Medium, 2026)

preset="stable_audio_3"

This instrumental music model generates an AI cover using the input audio as a reference. The settings let you make the result more or less similar to the original. Does not work with singing! Remix your instrumentals, turn noise into musical sounds, and repair low frequencies.

Parameters
model_nameModel variant

Allowed values

  • stable-audio-3-mediumStable Audio 3 MediumRecommended
strengthRemix Strength

High values produce results very different from the input audio. Sometimes too different.

Common values

  • 0.01Minimal edit
  • 0.1Light edit
  • 0.3Bit more edit
  • 0.5Balanced editRecommended
  • 0.6Moderate remix
  • 0.7Strong remix
  • 0.95Totally different song
promptPrompt

Use keywords to describe the style you want. For example: *italian western, cinematic score, mellow strings and choir*

prompt_strengthPrompt Strength

Increases the prompt's influence relative to the original music. High values produce the most creative and different results.

Common values

  • 0.5Less prompt strengthThe prompt is mostly ignored.
  • 1Base prompt strengthRecommendedBalanced.
  • 2More promptA little more prompt.
  • 3Stronger promptEven more prompt.
  • 4Even more promptOK, this is where it gets really fun. The model responds much better to the prompt.
  • 5High prompt strengthCareful: every comma in the prompt counts!
  • 6Very high prompt strengthWow! You might hear some entirely new sounds.
  • 7Maximum promptAt this level, you can use any input audio: the model follows only the prompt.
inpaint_regionsInpaint regions

AI Cover - YuE 2 (YuE2 3B, 2026)

preset="yue_2"

This music model detects melodies, harmonies, and sometimes even lyrics in the reference audio. It uses them to generate a high-quality cover. The result usually has a very different rhythm. The model loves to sing gibberish. Treat it as creative raw material.

Parameters
promptPrompt

Use keywords to describe the genre, instruments, language, voice...

instrumental_onlyInstrumental

Combines the detected vocal and instrumental melodies and gives the model no input for the vocal track. Unfortunately, the model usually adds gibberish vocals anyway.

prompt_strengthPrompt Strength

How strongly the cover follows your style prompt.

Common values

  • 0.1Minimal
  • 0.5Light
  • 11.0 - StandardRecommended
  • 1.2Strong

AI Cover - ACEStep 1.5 XL (ACE-Step 1.5 XL, 2026)

preset="acestep_15_xl"

This instrumental music model generates an AI cover using the input audio as a reference. Generations often sound very close to the reference unless you greatly reduce the original's influence. Recommended for improving bass tracks, reconstructing missing notes, and correcting specific artifacts.

Parameters
model_nameModel

Allowed values

  • acestep-v15-xl-turboAceSTEP turbo
  • acestep-v15-xl-sftAceSTEP SFT
strengthOriginal Influence

How much the original audio influences the result. 1.0 means the model tries to match the original as close as possible. 0.1 means it's a loose reference.

Common values

  • 1.0As close as possible
  • 0.99Almost original
  • 0.5Inspiration
  • 0.1Creative interpretation
promptPrompt

Use keywords to describe the style, instruments, tempo, and key.

High-Frequency Restoration - AudioSR (Audio SR, 2022)

preset="audiosr"

This model regenerates high frequencies above the selected cutoff and keeps the original audio below it. The lower the cutoff, the more of the sound it reconstructs. Great for historical recordings, poor-quality recordings, or recordings with unwanted high-frequency noise. It does not change the bass (use an AI cover model instead).

Parameters
frequency_cutoffStrength

Sets the frequency above which AudioSR regenerates the high end. A low cutoff reconstructs more of the frequency spectrum. A high cutoff focuses on the highest frequencies (which are less audible).

Allowed values

  • 16000Very Subtle (16 kHz cutoff)
  • 13000Subtle (13 kHz cutoff)
  • 10000Mild (10 kHz cutoff)
  • 8000Medium (8 kHz cutoff)Recommended
  • 5000Pronounced (5 kHz cutoff)
  • 4000Strong (4 kHz cutoff)
  • 3000Very Strong (3 kHz cutoff)
reconstruction_methodReconstruction

Allowed values

  • multiband_ensembleMultiband EnsembleRecommendedA gradual transition between the original low frequencies and the restored high frequencies.
  • original_signalContinuous ReconstructionA sharp cutoff between the original audio below the threshold and the restored high frequencies.
stereo_modeChannels

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.

High-Frequency Restoration - UniverSR (UniverSR, 2026)

preset="universr"

An upgraded version of AudioSR for music, voice, and sound effects. UniverSR reconstructs missing or poor-sounding high frequencies to add brightness. Try it on dull voices, old recordings (such as cassettes), or AI music from SUNO. It does not change the low frequencies (use an AI cover model instead).

Parameters
frequency_cutoffStrength

Allowed values

  • 12000Subtle (12 kHz cutoff)
  • 16000Very Subtle (16 kHz cutoff)
  • 8000Medium (8 kHz cutoff)Recommended
  • 6000Pronounced (6 kHz cutoff)
  • 4000Strong (4 kHz cutoff)
reconstruction_methodReconstruction

Allowed values

  • seamless_reconstruction_softSoft ReconstructionRecommendedRecommended. Attenuates the restored high frequencies for a softer sound. Fewer audible artifacts, but less brightness. Boost it with an EQ.
  • original_signalContinuous ReconstructionBlends restored high frequencies smoothly with the original audio.
  • originalAuthors' TechniqueThe original method from the research paper. This technique often leaves a visible transition in the spectrogram.
model_nameModel

Allowed values

  • universr-audioOriginalRecommended
  • universr-audio-finetune-v1Fine-tune v1
  • pitbull_edm_206Pitbull EDM
stereo_modeStereo

Allowed values

  • mid_sidesStereo (Mid/Sides)Process Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)RecommendedProcess the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.

Speech Enhancement - NovaSR Upscaler (NovaSR, 2026)

preset="novasr"

This upscaling model was trained on English speech. Use it to remove the walkie-talkie sound from your podcasts, voice-overs, or isolated vocal tracks.

Parameters
stereo_modeChannels

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.

High-Frequency Restoration - FlashSR (FlashSR, Jan 2025)

preset="flashsr"

Similar to AudioSR, this model reconstructs high frequencies more quickly. Handy for quickly upscaling audio from 44.1 kHz to 48 kHz. It does not change the low frequencies (use an AI cover model instead).

Parameters
stereo_modeStereo Mode

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.

Correct Tempo Variations (Constant BPM) (Beat This!, 2024)

preset="constant_bpm"

SUNO tracks often have an unstable tempo, subtly speeding up or slowing down throughout the music. This model detects the tempo and corrects these variations. Ideal before importing a track into a DAW or playing it in a DJ set. Also useful for live recordings. Works best when there are drums: EDM, rock, pop, techno, house, etc.

Parameters
quantize_precisionBeat quantization

Snap detected beats to the nearest musical subdivision.

Allowed values

  • 1/4
  • 1/8
  • 1/8T
  • 1/16Recommended
  • 1/16T
  • 1/32
  • 1/32T

Neural Reconstruction (DACVAE, 2024)

preset="dacvae"

This model passes the sound through a high-quality encoder-decoder model. The effect is very subtle: it can help correct certain artifacts (especially hi-hats in your drum'n'bass tracks).

Parameters
stereo_modeChannels

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.

Correct Clipping (Declip)

preset="declip"

Heavily compressed or limited audio sometimes has a waveform that is flat in places (clipping). This digital artifact, caused during recording or mixing, creates unpleasant noise. Use this model to correct it.

Parameters
model_nameModel

Allowed values

  • simple_declipSimple CorrectionRecommendedAutomatically corrects clipping while maintaining high perceived loudness.
  • betterdeclipperAdvanced CorrectionSeveral experimental mathematical techniques to remove clipping from very, very damaged audio.
modeMode

Auto mode is recommended.

Allowed values

  • autoAutoRecommendedAnalyzes flat clipping, smeared peaks from lossy encoding or resampling, limiter ceilings, and soft shoulders. Restores the detected damage and leaves unaffected regions intact.
  • hardHard clippingRestores detected clipping ceilings without a soft shoulder. At limiter ceilings, repairs only flat runs of two or more samples.
  • softSoft shoulderUses the same analysis as Auto but always restores a soft region below each ceiling, using the detected knee or 80% of the ceiling.
  • limiterLimiter (experimental)Restores a deeper soft region from 50% of limiter ceilings. Can reshape clean peaks and add audible distortion; compare carefully with Auto.
  • legacyLegacyUses the previous amplitude-histogram detection: repairs flat clipping plateaus, or a soft knee when no plateau is found.

Remove Limiting (Delimiter) (BS-RoFormer De-limit, becruily)

preset="delimiter"

Undo the brickwall limiting of an over-compressed master and bring back its punch and dynamics. In 16 or 24-bit, the result is 12 dB quieter to leave room for the restored peaks. Choose 32-bit float in the settings to keep the original level.

Noise Reduction

preset="denoise"

Remove hiss, hum, background noise, rumble, or animal and crowd sounds. Choose the variant based on the noise present and the sound you want to keep.

Parameters
model_nameModel

Allowed values

  • denoiseDenoiseRecommended
  • denoise_debleedDenoise and Debleed
  • custom_speechSpeech only
  • custom_music_singingMusic with singing
  • custom_musicMusic Only (no vocals)
  • custom_singingSinging only (no instruments)
promptSound to keep

Describe the sound to keep in the cleaned track.

stereo_modeStereo mode

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.
spansSound hints

Remove Reverb

preset="dereverb"

Get a dry sound from a recording drenched in reverb or echo, such as one recorded in a cellar or a church. Also useful for removing the slight echo of a poorly soundproofed room.

Parameters
model_nameModel

Allowed values

  • dereverbLong Reverb (strong)Recommended
  • dialogue_isolateRoom echo (light)

Crowd Noise Reduction

preset="decrowd"

Improve your live recordings! Keep the performance and remove audience noise: coughing, whistling, shouting, applause, etc. Use the Before/After slider to control the ambience level.

Parameters
model_nameModel

Allowed values

  • decrowdRoformerRecommended
  • decrowd_mdxMDX Net
  • custom_music_singingSAM Audio
promptSound to keep

Describe the sound to keep in the cleaned track.

stereo_modeStereo mode

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.
spansSound hints

Keep Only Center Mono (Gillian CenterWide v3, 2026)

preset="phantom_center"

This model keeps the mono center of a track and removes the content spread across the sides. Handy for isolating a bass, kick drum, or podcast voice. You can also use it to remove phaser, chorus, or flanger from an instrument.

Stereoize (Gillian Stereoizer Alpha 1)

preset="stereoize"

Convert mono audio into stereo by generating separate left and right channels. Stereo input is mixed to mono before processing.

Speech Enhancement - RE-USE Upscaler (NVIDIA RE-USE, 2026)

preset="reuse"

This model dramatically improves speech quality in multiple languages. It recovers high-frequency detail and reduces noise, reverb, and audio defects. For the most difficult restorations of podcasts, voice messages, or singing. Hallucinations are common: use UniverSR or blend with the original audio to improve the result.

Parameters
stereo_modeChannels

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.

Speech Bandwidth Extension - AP-BWE (AP-BWE, 2024)

preset="ap_bwe"

Makes muffled phone calls, voice messages, and low-quality voice recordings sound clear and full again.

Parameters
model_nameModel

Allowed values

  • ap-bwe-48k48 kHz upsamplingRecommended
  • ap-bwe-16k16 kHz upsampling
frequency_cutoffStrength

Sets the frequency above which AP-BWE rebuilds the voice. Pick the frequency where your recording's high end stops: a lower cutoff rebuilds more of the spectrum.

Allowed values

  • 12000Subtle (12 kHz cutoff)
  • 8000Medium (8 kHz cutoff)Recommended
  • 6000Pronounced (6 kHz cutoff)
  • 4000Strong (4 kHz cutoff)
frequency_cutoffStrength

Sets the frequency above which AP-BWE rebuilds the voice. Pick the frequency where your recording's high end stops: a lower cutoff rebuilds more of the spectrum.

Allowed values

  • 4000Strong (4 kHz cutoff)Recommended
  • 2000Very Strong (2 kHz cutoff)
  • 1000Extreme (1 kHz cutoff)
stereo_modeChannels

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.

MP3 Singing Restoration (Apollo Voice, 2025)

preset="apollo_voice"

An MP3 restoration model specialized in singing. Handy for improving a cappella vocals downloaded as MP3s from the internet or correcting certain compression artifacts.

Parameters
stereo_modeStereo Mode

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.

Deeffect Vocals (Gillian DeEffect v1 beta, 2026)

preset="deeffect_vocals"

Get a clean, dry lead vocal from a processed recording. Removes reverb, delay, vocal doubles, and backing vocals, and suppresses the instruments.

Mastering · Choose how the master should sound

Modern

genre="modern"

Balanced, clean mastering that preserves the character of your mix. A good starting point if you're not sure which style to choose.

Hip Hop

genre="hiphop"

Give the bass more weight and the drums more impact without losing vocal clarity. For hip-hop mixes where the bass lacks depth or the drums lack punch.

Techno

genre="techno"

Driving low end and crisp highs for club sound systems. Keep your track energetic and control the bass throughout.

Rock

genre="rock"

Bring out the guitars in the midrange with controlled saturation. Keep your rock mixes energetic and punchy.

R&B

genre="rnb"

A warmer, softer sound with rich low end and a clear vocal. For R&B mixes where the vocal should feel intimate and stay prominent.

Trap

genre="trap"

Heavy 808s, hard-hitting drums, and sharp attacks. Give your trap mixes more impact while keeping the deepest lows under control.

Mastering · Choose the target loudness

-8 LUFS

target_lufs="-8"

Maximum punch and energy for high-impact releases. Recommended for pop, electro, and metal tracks that need competitive loudness.

-9.5 LUFS

target_lufs="-9.5"

Competitive industry-standard loudness while retaining good audio quality. Recommended for rock, R&B, and rap.

-14 LUFS

target_lufs="-14"

Streaming-oriented loudness with more preserved dynamic range. Recommended for jazz, classical, funk, and music that should breathe.

Auto (-1 dBTP true peak, max LUFS)

target_lufs="auto"

Finds the loudest master that stays at or below -1 dBTP true peak instead of matching a fixed LUFS target.

Stem separation · Choose the stems you want

2 tracks

preset="2_tracks"

Separate the singing from the instrumental. Get a cappella vocals or a karaoke version for your remixes.

Parameters
model_nameModel variant

Allowed values

  • deuxDeux modelRecommended
  • leapLeap Xe

4 stems

preset="4_tracks"

Split the song into four tracks: bass, drums, vocals, and everything else.

6 stems

preset="6_tracks"

Split the song into six tracks: bass, drums, vocals, guitar, piano, and everything else.

53 stems

preset="mvsep_mega_53"

This model distinguishes up to 53 instrument tracks. Select the instruments present in your song. For the best quality, choose a specialized model when the instrument is available in the list.

Parameters
selected_stemsInstrument stems to keep

Allowed values

  • bassBass
  • bass_double-bassDouble Bass
  • drumsDrums
  • drums_congasCongas
  • drums_hhHi-Hats
  • drums_kickKick
  • drums_percussionPercussion
  • drums_snareSnare
  • drums_tambourineTambourine
  • drums_timpaniTimpani
  • drums_tomsToms
  • drums_triangleTriangle
  • guitarGuitar
  • guitar_acoustic-guitarAcoustic Guitar
  • guitar_dobroDobro
  • guitar_electric-guitarElectric Guitar
  • guitar_banjoBanjo
  • guitar_mandolinMandolin
  • guitar_sitarSitar
  • guitar_ukuleleUkulele
  • other_accordionAccordion
  • other_bellsBells
  • other_glockenspielGlockenspiel
  • other_marimbaMarimba
  • other_wind-chimesWind Chimes
  • pianoPiano
  • piano_digital-pianoDigital Piano
  • piano_harpsichordHarpsichord
  • piano_keysKeys
  • piano_organOrgan
  • piano_synthSynth
  • stringsStrings
  • strings_bowed_stringsBowed Strings
  • strings_celloCello
  • strings_harpHarp
  • strings_violaViola
  • strings_violinViolin
  • vocalsVocals
  • vocals_back-vocalBacking Vocal
  • vocals_lead-vocalLead Vocal
  • winds_bassoonBassoon
  • winds_brassBrass
  • winds_clarinetClarinet
  • winds_fluteFlute
  • winds_french-hornFrench Horn
  • winds_harmonicaHarmonica
  • winds_oboeOboe
  • winds_saxophoneSaxophone
  • winds_tromboneTrombone
  • winds_trumpetTrumpet
  • winds_tubaTuba
  • winds_windWind
  • winds_woodwindWoodwind
  • otherOther

Custom Prompt

preset="custom"

Describe the instrument or sound you want to isolate in the prompt.

Parameters
promptSound to isolate

Describe the instrument, voice, noise, or sound to isolate.

stereo_modeStereo handling
spansPrompt hint ranges

Speech Only

preset="custom_speech"

Keep the speech and remove background noise. For cleaning up recordings where the voice is the only sound you care about.

Parameters
promptSound prompt
stereo_modeStereo handling
spansPrompt hint ranges

Noise Reduction

preset="denoise"

Keep your music or voice and remove background noise. Choose the variant based on what you want to keep.

Parameters
promptSound to keep

Describe the sound to keep in the cleaned track.

stereo_modeStereo mode

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.
spansSound hints
model_nameModel variant

Allowed values

  • denoiseDenoiseRecommended
  • denoise_debleedDenoise and Debleed
  • custom_speechSpeech only
  • custom_music_singingMusic with singing
  • custom_musicMusic Only (no vocals)
  • custom_singingSinging only (no instruments)

Crowd Noise Reduction

preset="decrowd"

Quickly remove audience noise from your live recordings.

Parameters
promptSound to keep

Describe the sound to keep in the cleaned track.

stereo_modeStereo mode

Allowed values

  • mid_sidesStereo (Mid/Sides)RecommendedProcess Mid ((L + R) / 2) and Sides ((L - R) / 2) independently, then recombine them as stereo.
  • sides_onlySides onlyProcess the sides and keep the original center.
  • mid_onlyMid onlyProcess the center and keep the original sides.
  • left_rightStereo (L/R)Process the Left and Right channels independently.
  • monoMonoAverage Left and Right into one mono channel, then process it once.
spansSound hints
model_nameModel variant

Allowed values

  • decrowdRoformerRecommended
  • decrowd_mdxMDX Net
  • custom_music_singingSAM Audio

Enhanced Crowd Noise Reduction

preset="custom_music_singing"

SAM Audio Large keeps music and singing, and removes conversations, crowd noise, and other audience sounds.

Parameters
promptSound prompt
stereo_modeStereo handling
spansPrompt hint ranges

Music Only (No Vocals)

preset="custom_music"

Keep the instruments and remove singing, speech, and background noise.

Parameters
promptSound prompt
stereo_modeStereo handling
spansPrompt hint ranges

Singing Only

preset="custom_singing"

Keep the singing and remove instruments and ambient noise. Get isolated vocals to remix or process separately.

Parameters
promptSound prompt
stereo_modeStereo handling
spansPrompt hint ranges

Lead / Backing Vocals

preset="lead_back"

Separate the lead vocal from the backing vocals. Rework the lead singer without affecting the background voices.

Duet Singers

preset="duet"

Separate the two singers into two tracks. Rework each voice separately.

Main Singer

preset="custom_main_solo_singer"

Keep only the lead singer's voice. Backing vocals and other sounds are isolated separately.

Parameters
promptSound prompt
stereo_modeStereo handling
spansPrompt hint ranges

Isolate Drums

preset="2_tracks_drums"

A specialized model for separating drums from the rest of the song. You get two tracks: the drums and a version without drums. The model preserves detail in both.

5 drums tracks

preset="5_drums_tracks"

Split the drums into five tracks: kick, snare, toms, hi-hat, and cymbals.

Isolate Any Percussion

preset="custom_remove_drums"

Remove all percussion, including shakers, cymbals, and cinematic percussion.

Parameters
promptSound prompt
stereo_modeStereo handling
spansPrompt hint ranges

Remove Reverb

preset="dereverb"

Remove reverberation, delay, and echo from a track.

Keep Only Center Mono

preset="phantom_center"

This model keeps the mono center of a track and removes the content spread across the sides. Handy for isolating a bass, kick drum, or podcast voice. You can also use it to remove phaser, chorus, or flanger from an instrument.

Multichannel upmix

preset="5_1_upmix"

This model distributes the sound into a front stereo pair, a side stereo pair, and a center channel. The 5.1 version also adds a channel for sub-bass.

Parameters
model_nameModel variant

Allowed values

  • 5_1_upmix5.1 upmixRecommendedThe first upmix model, for film soundtracks.
  • 5_0_upmix_2026095.0 upmix JasperA newer upmix model for film soundtracks. Offers better quality than the first version.
  • 5_0_upmix_2026105.0 upmix GillianThis upmix model was trained only on music. Choose it for your songs.

Speech / Music / SFX

preset="speech_music_sfx"

Split a TV show, film, or video soundtrack into three tracks: speech, music, and sound effects. Remove the music from a video or keep only the dialogue.

Guitar

preset="guitar"

Isolate the guitar in one track. The rest of the song is kept in another.

Modern Bowed Strings

preset="modern_bowed_strings"

Isolate the bowed strings in one track and keep the rest of the song separate.

Synth Lead

preset="synth_lead"

Isolate the synth lead and keep the rest of the song in another track.

Transcription · Choose how the MIDI tracks are created

MuScriptor Large

model="muscriptor-large"

This model transcribes music into a MIDI file with multiple instruments. Its large training dataset and the computing power of the GPU it uses give it impressive performance. Note: this model does not generate note velocities.

Parameters
instrument_modeInstrument track mode

Allowed values

  • autoAutoRecommended
  • selectSelect instruments
  • combine_melodicCombine melodic instruments
instrumentsInstrument guidance

Allowed values

  • electric_bassBass
  • acoustic_bassAcoustic Bass
  • drumsDrums
  • acoustic_guitarAcoustic Guitar
  • clean_electric_guitarClean Electric Guitar
  • distorted_electric_guitarDistorted Electric Guitar
  • acoustic_pianoAcoustic Piano
  • electric_pianoElectric Piano
  • organOrgan
  • synth_leadSynth Lead
  • synth_padSynth Pad
  • synth_stringsSynth Strings
  • orchestra_hitOrchestra Hit
  • string_ensembleString Ensemble
  • violinViolin
  • violaViola
  • celloCello
  • contrabassContrabass
  • orchestral_harpOrchestral Harp
  • voiceVoice
  • chromatic_percussionXylophone
  • timpaniTimpani
  • brass_sectionBrass Section
  • trumpetTrumpet
  • tromboneTrombone
  • french_hornFrench Horn
  • tubaTuba
  • soprano_and_alto_saxSoprano & Alto Saxophone
  • tenor_saxTenor Saxophone
  • baritone_saxBaritone Saxophone
  • flutesFlutes
  • clarinetClarinet
  • oboeOboe
  • english_hornEnglish Horn
  • bassoonBassoon
quantize_precisionNote quantization

Allowed values

  • 1/4
  • 1/8
  • 1/8T
  • 1/16
  • 1/16T
  • 1/32
  • 1/32T

Transkun V2

model="transkun"

This high-quality model turns a piano performance into a MIDI track. It captures the notes, their velocities, and pedal controls. Magical for piano, but not useful for other instruments.

Parameters
quantize_precisionNote quantization

Allowed values

  • 1/4
  • 1/8
  • 1/8T
  • 1/16
  • 1/16T
  • 1/32
  • 1/32T

Code examples

Server-side example

JavaScript (Node.js)
const response = await fetch("https://api.neuralanalog.com/presets/run", {
  method: "POST",
  headers: {
    "X-API-Key": process.env.NEURALANALOG_API_KEY,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "operation": "restoration",
    "audio_id": "00000000-0000-0000-0000-000000000000",
    "preset": "preset"
  }),
});
if (!response.ok) {
  throw new Error(`Request failed with ${response.status}`);
}
console.log(await response.json());

Parameters

Send the API key from a trusted server. Never expose it in client-side JavaScript.

JSON request body

operation
requiredstring

Selects the preset family: restoration repairs or upscales audio, mastering controls loudness and tone, stem_split separates sounds, and transcription creates MIDI.

"restoration""mastering""stem_split""transcription"
audio_id
requiredstring

Parent imported audio asset ID.

preset
requiredstring

Preset identifier returned by GET /presets or the MCP catalog tool.

source_type
optionalstring

Selects a raw artifact or a rendered Studio mix as input. original, stem, restored, and mastered use the stored file without realtime mixer effects. current_main_mix renders the selected main-track version with its effects; current_stems_mix renders all audible stems with mute, solo, level, pan, and effects; current_stem_mix renders one stem's selected version and effects.

"original""stem""restored""mastered""current_main_mix""current_stems_mix""current_stem_mix"

Default: "original"

source_id
optionalstring | null

Object ID for stem, restored, mastered, or current_stem_mix. For current_stem_mix this is the stem ID. Omit it for original, current_main_mix, and current_stems_mix.

bit_depth
optionalinteger

Output WAV bit depth for audio-producing presets; 32 is float.

162432

Default: 24

parameters
optionalobject

Preset-specific values using only parameter keys returned for the selected preset. Mastering also accepts target_lufs.

Successful response

200Successful Response
object_type
requiredstring

No description provided.

"upscaled""mastered""stem_split""transcription""workflow"
object_id
requiredstring

No description provided.

audio_id
requiredstring

No description provided.

status
requiredstring

No description provided.

Errors

A missing or invalid X-API-Key returns an authentication error. Validation errors use the declared 422 response below.
422Validation Error
detail
optionalarray<object>

No description provided.

Next steps

Read the preset catalogPoll processing statusDownload the completed result