100%無料 — 登録不要 — データ収集なし

この曲は
AIが作った?

無料AI音楽検出ツール — Suno、Udio、Riffusionを数秒で検出

オーディオファイルをアップロードして数秒で判定を取得。ASCはAIジェネレーターが隠せない70以上の法医学的信号を分析。登録不要、カード不要。

オーディオファイルをここにドロップまたは ファイルを選択

MP3, WAV, FLAC, OGG, M4A — Max 50 MB

対応: YouTube、Spotify、SoundCloud

Song preview thumbnail

特徴量はローカル抽出。ASCスコアリングは安全なサーバー。

100%無料 登録不要 データ収集なし カード不要

3ステップ、10秒

Upload your file — our algorithm does the rest. Here's what happens behind the scenes.

01

任意のオーディオファイルをアップロード

Drag & drop an MP3, WAV, FLAC, OGG, or M4A file (up to 50 MB). The audio stays in your browser — we never store your music files.

02

深層オーディオ分析

Our engine scans 82+ forensic signals: frequency patterns, timing precision, spectral fingerprints, stereo behavior — things AI generators always leave behind.

03

AI vs 人間の判定

A proprietary scoring algorithm weighs every signal and returns a clear verdict: AI-generated, human-made, or a mix of both — plus which AI platform was used.

The Most Advanced Free AI Music Detector

AI Song Checker analyzes 82+ forensic signals to determine if a track was generated by artificial intelligence. Our proprietary engine identifies output from Suno, Udio, Riffusion, ElevenLabs, Stable Audio, MusicGen, Mureka, and other AI music generators. Includes C2PA & SynthID watermark detection.

7 AI Generators Detected

Platform-specific fingerprinting for Suno, Udio, Riffusion, ElevenLabs, Stable Audio, MusicGen (Meta), and Mureka. Each AI leaves unique audio signatures our engine recognizes.

C2PA & SynthID Watermarks

Detects digital watermarks embedded by Google SynthID and C2PA content credentials in audio files, adding another layer of verification beyond spectral analysis.

82+ Forensic Signals

Upload any MP3, WAV, FLAC, OGG, or M4A file. Our AI song checker extracts spectral fingerprints, MFCC coefficients, phase coherence patterns, and dozens more forensic signals directly in your browser.

Built In-House, Not a Wrapper

Our detection engine is developed from scratch — not a wrapper around someone else's API. Every line of the algorithm is ours. The more people use it, the smarter it gets.

Authenticity Certificates

Generate tamper-proof HMAC-SHA256 signed certificates for tracks verified as human-made. Trusted by musicians and labels screening submissions.

Public REST API

Integrate AI detection into your workflow. SDKs for Python, Node.js, and Java. Free tier: 100 requests/day. Chrome Extension for real-time detection on YouTube, SoundCloud, and Spotify.

Real-time visualization of the audio waveform during analysis.

Analyzing your track...

0%

Decoding audio file...

Your Analysis History

No checks yet. Upload a track to get started.

Source Track Date Result

プランを選択

無料アカウント

0€
  • 無制限分析
  • 基本結果 + 履歴
  • アカウントなしで1日3分析
  • 広告あり

よくある質問

Is it really free?

Yes, 100% free. 3 analyses per day without any signup. Unlimited with a free account (just an email). No credit card, no audio data stored on our servers. Unlike competitors, no paid subscription required.

How accurate is the AI Song Checker?

Our ASC v8.3 engine uses Bayesian scoring across 82+ forensic signals to achieve high accuracy on Suno, Udio, Riffusion, ElevenLabs, Stable Audio, MusicGen, and Mureka tracks.

Is my music uploaded to your servers?

Feature extraction happens locally in your browser. Only extracted features are sent to our secure scoring server. Audio files are stored encrypted if you're signed in.

What file formats are supported?

MP3, WAV, FLAC, OGG, and M4A files up to 50 MB.

Do I need an account?

No! Analyze 3 tracks per day without any account. For unlimited analyses and history, create a free account in 10 seconds (just an email). No credit card ever asked.

How does AI music detection work?

AI music has specific signatures: frequency cutoffs, high frame similarity, compressed dynamics, synthetic phase patterns, watermark patterns. Our proprietary engine measures 82+ forensic signals and detects C2PA/SynthID watermarks to detect these.

What makes you different from competitors?

Most AI music detectors require a paid subscription. AI Song Checker is free — and includes a Public REST API, Chrome Extension, Authenticity Certificates, URL Analysis, and C2PA/SynthID watermark detection. Our algorithm is built from scratch, no third-party API, and we don't store or resell your audio data.

What do you do with my data?

Audio analysis happens locally in your browser. Only extracted numerical features are sent to our scoring server. We don't store your audio files, we don't listen to them, and we don't sell any data. Your music stays yours.