Dynamics Processing
Dynamics processing controls volume variations in your audio. TNT analyzes peak levels, average energy, and dynamic range before processing. Designed for spoken content, but may work with music. The first two presets are relatively transparent; Broadcast preset applies aggressive multiband compression.
TNT uses Dynamic Scoring to determine compression characteristics. Select your compression style; the software determines exact values within each preset. Compression increases with higher processing tiers.
Off: No dynamics processing applied. Use when audio is already compressed or when original dynamics must be preserved.
Light: Gentle compression reducing only the loudest peaks. Identifies peak RMS levels and applies subtle compression (2.5:1 ratio). Attack and release times preserve transients while smoothing loud moments.
Appropriate for: well-recorded content needing minimal adjustment, acoustic music with intentional dynamics, content requiring natural dynamics.
Example: Music track with Dynamic Score 18.64 → 17.33 after Light processing.
Moderate: Standard broadcast compression for most content. Analyzes average RMS level and applies moderate compression (3.5:1 ratio). Calculates makeup gain automatically based on compression applied.
Works for: podcasts, voice-overs, most music, general broadcast material needing consistent playback across systems.
Example: Same track → Dynamic Score 14.58.
Broadcast: Aggressive multiband processing for maximum clarity and consistency. Splits audio into five frequency bands (sub-bass, bass, low-mid, mid, high) and analyzes each independently. Each band receives compression tailored to its characteristics—bass gets tighter control with longer attack times, highs receive faster compression for clarity.
Uses adaptive ratios: bass receives moderate compression (4.0:1), high frequencies get aggressive processing (up to 8.0:1). Attack and release times are frequency-dependent (200ms in bass, 100ms in highs).
Designed for: radio content, streaming platforms with varied playback systems, content consumed on small speakers or mobile devices, situations requiring maximum loudness and clarity.
Delivers varying results with music. Heavily compressed material (DS <9) likely causes distortion.
Example: Same track → Dynamic Score 12.71.
EQ Target Curves
EQ processing analyzes frequency response across ten octave-spaced bands (50Hz to 12.8kHz+). Measures RMS level, peak level, and crest factor for each band, then compares against professional target curves. Uses attenuation-focused philosophy—corrections are calculated, halved before application, with ±10 dB maximum adjustment. Designed for spoken content; delivers varying results with music.
Off: No equalization applied.
Flat: Targets pink noise curve with more energy in bass frequencies (-3 dB per octave rise from 1kHz). Attenuates frequencies exceeding this curve; leaves frequencies below curve unchanged. Prevents excessive energy buildup while maintaining natural tonal balance.
Appropriate for: well-balanced material, situations preventing frequency buildup without imposing specific tonal character.
Speech: Optimized for vocal clarity and intelligibility. Boosts presence frequencies (1.6kHz-6.4kHz) where consonants and speech intelligibility exist. Attenuates sub-bass rumble and reduces boxiness (200-400Hz range).
Designed for: podcasts, voice-overs, audiobooks, conference recordings, interviews, content requiring vocal clarity.
Broadcast: Aggressive clarity enhancement for small speakers, mobile devices, and varied listening environments. Emphasizes midrange intelligibility more strongly than Speech mode with deeper bass cuts and aggressive presence boost. Ensures content cuts through on phone speakers, laptop audio, and car radios.
Intended for: radio content, streaming platforms, mobile-first content, unknown playback systems, content requiring intelligibility on poor speakers.
Bypass All Processing
Disables both Dynamics and EQ processing regardless of selected settings. Use for loudness normalization only, without dynamics control or tonal shaping.
Useful for: testing audio with normalization alone, A/B comparing processed versus unprocessed versions, situations where processing is already applied in your DAW and only format conversion and loudness compliance are needed.
Processing Order
When multiple processing stages are enabled, TNT applies them in this order:
- EQ adjustments (if enabled)
- De-esser (automatically applied when EQ is active)
- Dynamic normalization
- Dynamics processing (if enabled)
- Loudness normalization (if enabled)
This signal chain ensures frequency balance is corrected before dynamics processing, preventing the compressor from reacting to frequency imbalances. De-esser removes harsh sibilance after EQ boosts but before compression, ensuring the compressor doesn't overreact to sibilant sounds. Loudness normalization happens last, after all processing is complete, guaranteeing target LUFS level is achieved accurately.
Notes
All processing happens at 192kHz sample rate internally for intersample peak accuracy. For 16-bit PCM output, triangular dithering is applied after all processing to minimize quantization artifacts. Multiband processing uses linear-phase crossover filters to prevent phase distortion between frequency bands.
TNT's adaptive processing means identical preset selections may produce different filter parameters depending on input audio characteristics. The software adjusts processing based on measurements, ensuring optimal results for each file rather than applying static presets.
