MixTester was built to solve a specific problem in the audio engineering world: bias.
When you listen to your own mixes, you know exactly what you did. You know you boosted the vocal by 1dB, or compressed the snare harder. That knowledge colors your perception.
We built MixTester to provide an objective, level-matched environment where the only thing that matters is the sound. By allowing gapless A/B switching and blind testing, we help engineers and producers make decisions based on their ears, not their egos.
Critical listening is fragile. A pause of even a few seconds between two audio files is enough for your auditory memory to fade. MixTester's gapless engine ensures that you can switch instantly, revealing subtle differences that would otherwise be lost.
Unlike standard HTML5 audio players which stream files individually, MixTester leverages the browser's Web Audio API. When you load a project, we decode and load the audio data into a shared buffer. This allows us to perform sample-accurate switching between sources without any buffering or "hiccups."
The "Loudness War" is real. Human hearing naturally prefers louder sounds, often mistaking them for "better." To combat this, MixTester includes an integrated LUFS (Loudness Units Full Scale) analyzer.
We analyze the integrated loudness of every track upon upload. When you enable Normalize Mode, we apply non-destructive gain compensation to match all tracks to the target level of the quietest track. This ensures you are comparing the mix balance, not the mastering limiter settings.
Files are uploaded to secure, transient object storage (R2). We generate unique, cryptographically random IDs for every project. We do not train AI models on your audio, and we do not claim ownership of any uploaded work. All files are automatically permanently deleted after 30 days to protect your intellectual property.