Welcome to the official documentation for the Sonodit Audio Pipeline. Our platform takes raw voiceover files and transforms them into broadcast-ready, mastered assets—completely hands-free. Here’s the step-by-step breakdown of how your audio travels through our intelligent engine:

Step 1: Ingest and Acoustic Profiling

As soon as your file hits the system, the Sonodit Audio Profiler performs a deep-signal X-ray. It measures sample rate, bit depth, and calculates the initial Crest Factor. Simultaneously, our neural network maps the voiceover’s phonetic structure to determine their exact Words Per Minute (WPM), native language, and regional dialect. This creates a predictive roadmap that guides every tool down the line.

Step 2: Subsonic Filtering and Noise Subtraction

To ensure total clarity, the audio enters the restoration phase. A dynamic High-Pass Filter is set at 50 Hz to scrub out those muddy low-end rumble frequencies. At the same time, the engine samples the 'noise DNA' detected during the speaker's pauses and applies a 50% inverse spectral subtraction algorithm, wiping out room noise without compromising the natural tone of the voice.

Step 3: Block Stabilization (Clip Gain Matching)

This is where things get technical: the audio undergoes dual-stage normalization. First, the entire signal is standardized to -25.0 dB RMS. Immediately after, the system analyzes the audio phrase-by-phrase. Whether the talent whispered a line or shouted a tagline, the algorithm calculates the exact volume delta and levels it out, applying a safety limit of +/-15.0 dB while protecting short pauses with an exclusion window. This ensures every word has the same punch and commercial presence.

Step 4: Dynamic Isolation (Predictive Gating)

The stabilized audio runs through an ultra-precise noise gate with a threshold between -50 dBFS and -55 dBFS. The system detects the voice’s attack transients to open the signal path instantly, applying a controlled fade on word tails to bury any unwanted room air or heavy breaths in dead silence.

Step 5: Acoustic Modeling and Commercial Texturing

Your clean file is fed into our virtualized DSP engine. Here, the audio passes through a simulated analog rack that applies soft-knee compression, tight de-essing, and parallel harmonic excitation. This phase glues the signal together, giving the voice that 'Big and Clear' broadcast sound you hear on top-tier radio and cinema spots, with a hard True Peak ceiling of -2.0 dB.

Step 6: Precision Segmentation and PTK Shielding

In the final stage, our 'Cartographer' (the temporal orchestrator) takes over. If you’re working in automated mode, it finds the silent valleys and makes clean cuts to export individual phrase files. Each cut includes a 10ms preemptive attack margin—what we call PTK Protection—ensuring no initial consonants get clipped. If you need a unified master, the system stitches the phrases together, injecting dynamic silences calculated based on the talent's natural rhythm to deliver a punchy, professional, and ready-to-sell piece.