/presetsChoose the right audio-processing preset
Read one canonical catalog for every audio-processing operation. Cache it in your client, filter the relevant preset family, and send the selected preset to POST /presets/run.
How this endpoint works
Use this endpoint before presenting processing choices or translating a natural-language request into a preset. The catalog includes defaults, recommended use cases, supported parameters, model variants, and quota multipliers. It is public and cacheable; do not fetch it before every job.
- 1
Read and cache the catalog
Fetch the catalog once and respect its cache headers instead of embedding a duplicate preset list in your app.
- 2
Filter by operation
Use restoration_presets, mastering_presets, stem_split_presets, or transcription_presets for the requested capability.
- 3
Run the selected preset
Run a preset by passing its identifier and only its advertised parameters to the unified processing endpoint.
Common use cases
Choose this operation when it matches the source and result your pipeline needs.
Route an AI request
Match requests such as clean this up [Suno link], remove the crowd [live.wav], master this [mix.wav], or turn this into MIDI [piano.mp3] to the relevant preset family.
Build processing controls
Populate selectors and parameter fields from the server instead of maintaining frontend constants.
Keep model choices current
Receive new presets and updated recommendations without shipping a client release.
Processing used by 80,000+ music makers
Results people rely on
“It can bring back clarity without making everything sound too artificial or overprocessed. Stem separation is also surprisingly clean.”
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.
MP3 and CD audio restoration
“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
“Easy to use and high quality results.”
Bjark
Choose a preset for the problem
Choose the preset that matches the problem you hear. Start with its default settings, then adjust the strength after listening.
| Operation / preset | Best suited to | Default settings |
|---|---|---|
restoration/universal_enhancer | MP3 compression; holes in the spectrogram; missing high frequencies above 16 kHz | stereo_mode="mid_sides" |
restoration/stable_audio_3 | 5 kHz ringing in instrumentals; instrumentals that need new sounds or musical ideas | model_name="stable-audio-3-medium", strength="0.5", prompt="high quality studio recording CD quality", prompt_strength="1" |
restoration/yue_2 | This music model detects melodies, harmonies, and sometimes even lyrics in the reference audio. | prompt="English, jazz-funk, warm lead vocal, Rhodes, bass and drums", instrumental_only="false", prompt_strength="1" |
restoration/acestep_15_xl | music or stems to remix with new notes and sounds | prompt="high quality studio recording CD quality" |
restoration/audiosr | hiss around 10 kHz in hi-hats or voices; high-end reconstruction above a selected cutoff | frequency_cutoff="8000", reconstruction_method="multiband_ensemble", stereo_mode="mid_sides" |
restoration/universr | missing high frequencies; too much high end or harsh hi-hats in AI audio; unnatural-sounding voices | frequency_cutoff="8000", reconstruction_method="seamless_reconstruction_soft", model_name="universr-audio", stereo_mode="left_right" |
restoration/novasr | high-frequency reconstruction with the English-speech model | stereo_mode="mid_sides" |
restoration/flashsr | missing high-end detail; single-instrument upscaling; preparing 44.1 kHz stems for Atmos export | stereo_mode="mid_sides" |
restoration/constant_bpm | unwanted changes in tempo; beats that drift off a fixed grid | quantize_precision="1/16" |
restoration/dacvae | rough or odd sound textures | stereo_mode="mid_sides" |
restoration/declip | clipped peaks; crackling caused by clipping | model_name="simple_declip", mode="auto" |
restoration/delimiter | brickwall limiting; squashed transients and dynamics | Preset defaults |
restoration/denoise | hiss; hum; background noise | model_name="denoise", stereo_mode="mid_sides" |
restoration/dereverb | long reverb tails; echo in music and vocals | model_name="dereverb" |
restoration/decrowd | crowd noise; cheering; applause | model_name="decrowd", stereo_mode="mid_sides" |
restoration/phantom_center | wide stereo sound around a centered voice, bass, or kick | Preset defaults |
restoration/stereoize | mono audio that needs stereo width | Preset defaults |
restoration/reuse | noisy speech; muffled voice detail | stereo_mode="mid_sides" |
restoration/ap_bwe | narrowband speech from phone calls or voice messages; voices missing high frequencies above a known cutoff | model_name="ap-bwe-48k", frequency_cutoff="8000", stereo_mode="mid_sides" |
restoration/apollo_voice | compression and missing detail in singing | stereo_mode="mid_sides" |
restoration/deeffect_vocals | reverb and delay on vocals; vocal doubles and backing-vocal bleed | Preset defaults |
mastering/modern | uneven tone or low loudness in a finished mix | target_lufs=-9.5 |
mastering/hiphop | light bass; soft drums; buried vocals in hip hop | target_lufs=-9.5 |
mastering/techno | loose low end; dull highs in electronic music | target_lufs=-9.5 |
mastering/rock | recessed guitars; drums that need more punch | target_lufs=-9.5 |
mastering/rnb | thin tone; vocals that need warmth and presence | target_lufs=-9.5 |
mastering/trap | uneven 808 bass; drum hits that need more definition | target_lufs=-9.5 |
stem_split/2_tracks | Separate the singing from the instrumental. | Preset defaults |
stem_split/4_tracks | Split the song into four tracks: bass, drums, vocals, and everything else. | Preset defaults |
stem_split/6_tracks | Split the song into six tracks: bass, drums, vocals, guitar, piano, and everything else. | Preset defaults |
stem_split/mvsep_mega_53 | This model distinguishes up to 53 instrument tracks. | selected_stems=["drums", "vocals", "bass", "piano", "strings", "piano_synth", "guitar", "other"] |
stem_split/custom | Describe the instrument or sound you want to isolate in the prompt. | Preset defaults |
stem_split/custom_speech | Keep the speech and remove background noise. | Preset defaults |
stem_split/denoise | Keep your music or voice and remove background noise. | stereo_mode="mid_sides" |
stem_split/decrowd | Quickly remove audience noise from your live recordings. | stereo_mode="mid_sides" |
stem_split/custom_music_singing | SAM Audio Large keeps music and singing, and removes conversations, crowd noise, and other audience sounds. | Preset defaults |
stem_split/custom_music | Keep the instruments and remove singing, speech, and background noise. | Preset defaults |
stem_split/custom_singing | Keep the singing and remove instruments and ambient noise. | Preset defaults |
stem_split/lead_back | Separate the lead vocal from the backing vocals. | Preset defaults |
stem_split/duet | Separate the two singers into two tracks. | Preset defaults |
stem_split/custom_main_solo_singer | Keep only the lead singer's voice. | Preset defaults |
stem_split/2_tracks_drums | A specialized model for separating drums from the rest of the song. | Preset defaults |
stem_split/5_drums_tracks | Split the drums into five tracks: kick, snare, toms, hi-hat, and cymbals. | Preset defaults |
stem_split/custom_remove_drums | Remove all percussion, including shakers, cymbals, and cinematic percussion. | Preset defaults |
stem_split/dereverb | Remove reverberation, delay, and echo from a track. | Preset defaults |
stem_split/phantom_center | This model keeps the mono center of a track and removes the content spread across the sides. | Preset defaults |
stem_split/5_1_upmix | This model distributes the sound into a front stereo pair, a side stereo pair, and a center channel. | Preset defaults |
stem_split/speech_music_sfx | Split a TV show, film, or video soundtrack into three tracks: speech, music, and sound effects. | Preset defaults |
stem_split/guitar | Isolate the guitar in one track. | Preset defaults |
stem_split/modern_bowed_strings | Isolate the bowed strings in one track and keep the rest of the song separate. | Preset defaults |
stem_split/synth_lead | Isolate the synth lead and keep the rest of the song in another track. | Preset defaults |
transcription/muscriptor-large | This model transcribes music into a MIDI file with multiple instruments. | instrument_mode="auto", instruments=["electric_bass", "drums", "acoustic_piano"] |
transcription/transkun | This high-quality model turns a piano performance into a MIDI track. | Preset defaults |
Lower frequency cutoffs rebuild more of the sound. Stable Audio 3 and ACEStep are remixes and can change musical content.
For several problems, remove noise or echo first, restore missing detail, then master the result. Use constant BPM for unwanted tempo drift and Declip for clipped peaks. Use the workflow catalog for ready-made sequences.
Code examples
Server-side example
const response = await fetch("https://api.neuralanalog.com/presets?locale=en", {
method: "GET",
headers: {
"X-API-Key": process.env.NEURALANALOG_API_KEY,
},
});
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.
Path and query parameters
localeSuccessful response
restoration_presetsNo description provided.
mastering_targetsNo description provided.
mastering_presetsNo description provided.
stem_split_groupsNo description provided.
stem_split_presetsNo description provided.
transcription_presetsNo description provided.
Errors
X-API-Key returns an authentication error. Validation errors use the declared 422 response below.detailNo description provided.