Interface AudioProjectParams

interface AudioProjectParams {
    bpm?: number;
    composerMode?: boolean;
    creativity?: number;
    disableNSFWFilter?: boolean;
    duration?: number;
    guidance?: number;
    keyscale?: string;
    language?: string;
    loras?: string[];
    loraStrengths?: number[];
    lyrics?: string;
    modelId: string;
    negativePrompt?: string;
    network?: SupernetType;
    numberOfMedia: number;
    outputFormat?: AudioOutputFormat;
    positivePrompt: string;
    promptStrength?: number;
    sampler?: string;
    scheduler?: string;
    seed?: number;
    shift?: number;
    steps?: number;
    stylePrompt?: string;
    timesignature?: string;
    tokenType?: TokenType;
    type: "audio";
}

Hierarchy (View Summary)

Properties

bpm?: number

Beats per minute (30-300, default: 120)

composerMode?: boolean

Enable AI composer mode for higher quality music generation (default: true). Disable for faster generation or when using reference audio. Maps to generate_audio_codes in the ComfyUI workflow.

creativity?: number

Composition variation / temperature (0-2, default: 0.85). Higher = more creative, lower = more predictable. Maps to temperature in the ComfyUI workflow.

disableNSFWFilter?: boolean

Disable NSFW filter for Project. Default is false, meaning NSFW filter is enabled. If image triggers NSFW filter, it will not be available for download.

duration?: number

Duration of the audio in seconds (10-600, default: 30)

guidance?: number

Guidance scale. For most Stable Diffusion models, optimal value is 7.5. For video models: Regular models range 0.7-8.0, LoRA version (lightx2v) range 0.7-1.6, step 0.01. This maps to guidanceScale in the keyFrame for both image and video models.

keyscale?: string

Key/scale setting (e.g., "C major", "A minor"). Omitted to use server default.

language?: string

Lyrics language code (default: en)

loras?: string[]

Array of LoRA IDs to apply. Available LoRAs are model-specific. The worker will download the LoRA if not already present on the persistent volume. LoRA IDs are resolved to filenames via the worker config API. Example: ['multiple_angles']

loraStrengths?: number[]

Array of LoRA strengths corresponding to each LoRA in the loras array. Values should be between 0.0 and 2.0. Defaults to 1.0 if not specified. Example: [0.9]

lyrics?: string

Song lyrics. Omit for instrumental generation.

modelId: string

ID of the model to use, available models are available in the availableModels property of the ProjectsApi instance.

negativePrompt?: string

Prompt for what to be avoided. If not provided, server default is used.

network?: SupernetType

Override current network type. Default value can be read from sogni.account.currentAccount.network

numberOfMedia: number

Number of media files to generate. Depending on project type, this can be number of images or number of videos.

outputFormat?: AudioOutputFormat

Output audio format. Can be 'mp3', 'flac', or 'wav'. Defaults to 'mp3'.

positivePrompt: string

Prompt for what to be created

promptStrength?: number

How closely the AI composer follows your prompt (0-10, default: 2.0). Higher values = stricter prompt adherence. Maps to cfg_scale in the ComfyUI workflow.

sampler?: string

Sampler, available options depend on the model.

scheduler?: string

Scheduler, available options depend on the model.

seed?: number

Seed for one of images in project. Other will get random seed. Must be Uint32

shift?: number

Shift parameter for ModelSamplingAuraFlow (1-6, default: 3 for turbo). Controls how denoising effort is distributed across generation steps. Higher values front-load structure/composition, producing more coherent arrangements. Lower values distribute effort evenly, focusing more on detail/texture. Official ComfyUI template uses shift=3 for ACE-Step 1.5 Turbo.

steps?: number

Number of steps. For most Stable Diffusion models, optimal value is 20.

stylePrompt?: string

Image style prompt. If not provided, server default is used.

timesignature?: string

Time signature (2, 3, 4, or 6 - default: 4)

tokenType?: TokenType

Select which tokens to use for the project. If not specified, the Sogni token will be used.

type: "audio"