SongUp AI
Documentation

#Overview

SongUp AI is a comprehensive AI-powered creative platform that enables anyone to create professional-quality music, images, videos, and speech using artificial intelligence. The platform features 8 dedicated creative studios and 18+ specialized audio tools — all accessible through a modern web interface at songupai.com .

SongUp uses cutting-edge proprietary AI engines (Sonic Flow, Neural Studio, Elite Master) along with Google Imagen, Veo, and Chirp3-HD to deliver studio-quality output. No prior experience in music production, audio engineering, image design, or video editing is required.

#Platform Capabilities

#Getting Started

Follow these steps to start creating with SongUp AI:

1. 1Visit songupai.com and sign in with your Google account. No credit card required — every new account starts with free credits.
2. 2Navigate to the main Studio page. You'll see a text prompt field at the center of the screen.
3. 3Type a description of the song you want to create. Be specific about mood, genre, tempo, and instruments for best results. Example: "upbeat pop song about summer with electric guitar and synth, female vocals".
4. 4Choose your Music Type (Vocal Song, Instrumental, Lo-Fi, etc.), Music Style (Pop, Rock, Jazz, etc.), and Duration.
5. 5Select an AI Engine: Sonic Flow (fast creative generation), Neural Studio (pro composition), or Elite Master (ultra studio quality).
6. 6Click Generate and wait while the AI creates your song. Generation typically takes 30–120 seconds depending on duration and engine.
7. 7Listen to the result, download it, or regenerate with tweaked parameters. All generated media is saved to your Library.

#AI Music Generation

The main AI Music Studio is the core of SongUp. It takes a text prompt and optional lyrics input and generates a complete, studio-quality song. Access it at /studio .

#Input Modes

SongUp supports five ways to create music:

#Music Types

#Music Styles

Choose from 17 predefined genres or leave as Default for AI to decide based on your prompt:

#Duration Options

#Negative Prompts

You can provide a negative prompt to specify what the AI should avoid in the output. Examples: "no drums", "no vocals", "avoid heavy bass", "no electronic sounds". This feature is available for all music engines.

#Humming to Song

Convert any hummed melody, whistled tune, or sung phrase into a fully produced song with AI. Access at /studio/humming .

#Input Methods

• Upload File — MP3, WAV, M4A, or OGG audio file (maximum 7 MB)
• Record Live — Use your microphone to record a melody in real-time with a visual timer
• Playback Preview — Listen to your recording before submitting for generation

#Generation Modes

How it works: SongUp uses the Basic Pitch machine learning model (TensorFlow.js) to extract MIDI note data from your melody recording. The detected notes, pitch, and rhythm are then passed to SongUp Neural Studio to compose a full arrangement. The model runs entirely in your browser with GPU acceleration support and caches model weights in IndexedDB for faster subsequent use.

Optional parameters include lyrics input and style selection from 15 genres.

#AI Image Studio

Generate stunning AI images, album artwork, and creative visuals from text descriptions. Powered by Google Imagen. Access at /studio/image .

1. 1Enter a detailed text prompt describing the image you want to generate.
2. 2Select your preferred aspect ratio — 1:1 for album covers, 16:9 for banners, 9:16 for mobile wallpapers.
3. 3Optionally set a seed number for reproducibility.
4. 4Click Generate. Processing typically takes 10–30 seconds.
5. 5Download the result in your preferred format (PNG, JPEG, or WebP) or create a new image.

#AI Video Studio

Generate cinematic AI videos from text descriptions. Powered by Google Veo. Access at /studio/video .

Video generation takes longer than other media types (typically 60–180 seconds). SongUp automatically resumes incomplete video generations if you navigate away and return. A cancel button appears after 10 minutes of processing.

#AI Speech Studio

Convert text to natural-sounding speech with AI voices. Powered by Google Chirp3-HD. Access at /studio/speech .

#Available Voices

#Background Music Generator

Generate AI soundtrack for any YouTube video. Paste a YouTube URL and the AI analyzes the video content to compose a matching background music track. Access at /studio/bgm .

1. 1Paste a valid YouTube video URL into the input field.
2. 2Click Generate. The AI will analyze the video content, mood, and pacing.
3. 3SongUp processes the video in three stages: Analyzing video content → Composing AI track → Saving to library.
4. 4The resulting song is saved to your library with mood, tempo, and instrument analysis metadata.

#Recording Studio

A full-featured in-browser recording studio with multi-track recording, real-time effects processing, and professional mixing controls. Access at /studio/recording .

#Track System

• Vocal Track — Record vocals directly from your microphone with real-time monitoring
• Instrumental Track — Import a backing track or instrumental file to record over
• Raw Mic Stream — Direct high-fidelity microphone capture without processing
• Metronome — Built-in metronome with adjustable tempo and volume

#Audio Effects

#Effect Presets

#Mixing Controls

• Individual track volume (0–150%)
• Pan control (-1 left to +1 right)
• Solo and Mute per track
• Master gain control
• Playback speed adjustment
• Export to WAV

#AI Chat Assistant

A conversational AI assistant specialized in music creation, lyrics writing, production advice, and creative brainstorming. Access at /studio/chat .

Chat history is preserved across sessions with conversation threading. You can start new conversations or continue existing ones.

#Audio Tools

SongUp includes 18+ specialized audio processing tools. All tools run directly in your browser using Web Workers, ONNX Runtime, and TensorFlow.js for fast, private processing. No audio is sent to external servers for tool processing (AI generation features use cloud APIs).

#Stem Separator

Split any song into individual stems using AI-powered source separation (HTDemucs v4, 172 MB ONNX model). Access at /tools/stem-separator .

Per-stem controls: Volume fader (0–150%), pan (-1 to +1), mute/solo toggle. Built-in presets include Mute Drums, Mute Bass, Mute Vocals, and Karaoke mode. Download the final mix as WAV with adjustable playback speed (0.5x–2x).

#AI Mastering

Professional audio mastering powered by AI. Upload any audio file and get loudness-normalized, EQ-optimized, dynamics-enhanced output ready for streaming platforms. Access at /tools/ai-mastering .

The mastering engine automatically applies loudness normalization (LUFS targeting), EQ optimization, dynamic range processing, and stereo enhancement. No manual settings needed — the AI analyzes your track and applies appropriate processing.

#Karaoke Maker

Remove vocals from any song and create a karaoke-ready instrumental track with synchronized lyrics display. Access at /tools/karaoke-maker .

Uses the same Demucs model as Stem Separator to isolate and remove vocals. The remaining instrumental mix is combined and synchronized with auto-transcribed lyrics for karaoke playback.

#Noise Remover

Clean audio recordings by removing background noise, hiss, hum, and unwanted ambient sounds. Access at /tools/noise-remover .

Ideal for cleaning up podcasts, voice recordings, interviews, and field recordings. Upload your noisy audio and download the cleaned version.

#Acapella Extractor

Isolate and extract clean vocals (acapella) from any song. Access at /tools/acapella-extractor .

Uses AI source separation to extract the vocal track while removing all instrumental elements. Perfect for creating remixes, vocal samples, covers, and mashups.

#Lyric Transcriber

Automatically transcribe lyrics from any song with word-level timing. Access at /tools/lyric-transcriber .

#Output Formats

Download all formats as a ZIP file or copy individual formats to clipboard. Word grouping splits on 8+ words, punctuation, or ~1.5 second silence gaps.

#Chord Recognition

Detect chords, BPM, and musical key from any audio file. Access at /tools/chord-recognition .

Outputs include detected BPM, musical key, time-based chord changes with confidence scores, and a color-coded chord timeline grouped by root note. Uses the Basic Pitch model for pitch analysis.

#Audio to MIDI

Convert any audio recording into editable MIDI data. Access at /tools/audio-to-midi .

Uses the Basic Pitch TensorFlow.js model for pitch detection. Outputs a downloadable MIDI file with note count statistics and duration tracking. You can play back the generated MIDI directly in the browser. The model runs client-side with GPU acceleration and is cached in IndexedDB.

#Smart Auto Mix

Intelligently mix multi-track audio recordings with AI-powered EQ, compression, panning, and level balancing. Access at /tools/smart-auto-mix .

Upload multiple stems and the AI automatically balances volumes, applies EQ, and creates a professional-sounding mix. No audio engineering experience needed.

#AI Blend

Merge and blend multiple audio tracks into seamless compositions with smart crossfading. Access at /tools/ai-blend .

#Neural Auto-Tune

AI-powered pitch correction that fixes vocal pitch issues while preserving natural vocal timbre. Access at /tools/neural-auto-tune .

Includes multiple styling presets and a graphical pitch visualization showing before/after comparison.

#Timbre Transfer

Transform the sonic character of any audio into a different instrument. Access at /tools/timbre-transfer .

#Target Instruments

Processing: Basic Pitch melody extraction → DDSP neural synthesis with the target instrument timbre → WAV export.

#Live Voice Changer

Real-time AI voice effects for streaming, calls, and content creation. Access at /tools/live-voice-changer .

Supports file upload (WAV, MP3, OGG) or live microphone input. Multiple voice presets available with custom pitch adjustment (-12 to +12 semitones) and formant control. Includes before/after playback comparison. All processing happens in the browser via Web Audio API.

#Live Karaoke

Sing karaoke with real-time synchronized lyrics, word-by-word highlighting, and live microphone blending. Access at /tools/live-karaoke .

• Upload audio or record from microphone
• Real-time word-by-word lyric highlighting
• Playback rate control
• Fullscreen mode
• Live microphone blending with backing track volume control
• GPU acceleration for smooth performance

#Sampling Tool

Extract and catalog audio samples, loops, and one-shots from any audio file. Access at /tools/sampling-tool .

Automatically detects sample-worthy segments and catalogs them for creative reuse in your productions.

#Mashup Creator

Blend multiple songs together with automatic BPM matching, key detection, and seamless transitions. Access at /tools/mashup-creator .

#Voice Swap Prep

Prepare audio tracks for AI voice conversion. Cleans, isolates, and optimizes vocals for the best voice swap results. Access at /tools/voice-swap-prep .

#Neural Synth

An AI-powered synthesizer with a piano keyboard interface, MIDI support, and neural sound generation. Access at /tools/neural-synth .

Supports multiple instrument presets and real-time playback. Computer keyboard keys are mapped to piano notes for playing without a MIDI controller.

#AI Engines & Models

SongUp AI uses multiple specialized AI engines for different media types:

#Music Generation Engines

#Other AI Models

#Pricing & Plans

SongUp AI offers five subscription tiers with three billing cycles. All plans include access to all studios and tools — the difference is credit allocation and processing priority.

#Plan Feature Comparison

Biannual billing saves 10% and annual billing saves 20% compared to monthly pricing. Payments are processed securely through LemonSqueezy. View current pricing at /pricing .

#Credits System

SongUp uses a credit-based system to meter AI generation usage. Credits are consumed when you generate music, images, videos, or speech. Audio processing tools (stem separator, mastering, etc.) that run client-side do not consume credits.

Your credit balance is displayed in the interface and refreshes each billing cycle based on your plan. Unused credits do not roll over. View your detailed credit breakdown at /credit-breakdown .

#Account & Settings

Manage your account, preferences, billing, and data from the Settings page at /settings . Settings are organized into six tabs:

#Social & Community

SongUp includes a built-in social platform for sharing and discovering AI-generated music. Access the community feed at /feed .

• Community Feed — Browse AI-generated songs shared by other creators
• Follow System — Follow your favorite creators and get updates when they post new content
• Like & Share — Heart songs you enjoy and share them via direct links
• Creator Profiles — View any creator's public profile and their shared songs
• Notifications — Get notified when someone follows you, likes your song, or shares your content
• Rating System — Rate and review AI-generated content
• Public Sharing — Generate shareable links for any of your creations

#Library & Downloads

All generated content (songs, images, videos, speech) is automatically saved to your Library at /library .

• Filter by type — Songs, Images, Videos, or Speech
• Playback — Listen to songs and speech directly in the browser with the built-in player
• Download — Download individual files or batch export as ZIP
• Share — Create public shareable links for any item
• Pin — Pin favorite items to the top of your library
• Delete — Remove individual items or use Bulk Delete in Settings
• In-Progress Tracking — View and monitor currently generating items via the sidebar

Download quality depends on your plan: Basic gets 128kbps MP3, Starter gets 320kbps MP3, and Pro+ gets lossless WAV exports.

#Supported Languages

SongUp AI's interface is fully translated into 30 languages. Change your language in Settings → Preferences.

AI-generated lyrics and song content can be created in any language supported by the underlying AI models — write your prompt or lyrics in your preferred language.

#Keyboard Shortcuts

Neural Synth piano keyboard mapping (computer keyboard → piano notes):

#API Reference

SongUp AI exposes internal API endpoints for all platform operations. These are used by the web client and are documented here for reference:

#Supported Formats

#Audio Formats

#Image Formats

#Video Formats

#Frequently Asked Questions