How audio encoders use psychoacoustic masking to reduce file sizes, and why this process creates audible artifacts in compressed formats.
Understanding what is loudness, how it's measured and why standards exist.
Understanding bit depth, quantization and why float sample rates are needed.
How randomization helps to alleviate the effects of quantization.
Please note that these options are available to some users only after 5th December 2025.
Dynamics processing controls how TNT manages the volume variations in your audio. The software analyzes peak levels, average energy, and dynamic range before applying any processing.
Off No dynamics processing is applied. Use this when your audio is already properly compressed or when you need the original dynamics preserved.
Light Gentle compression that reduces only the loudest peaks. The software identifies peak RMS levels and applies subtle compression with a 2.5:1 ratio. Attack and release times are set to preserve transients while smoothing out occasional loud moments. This preset maintains the natural character of your audio while preventing clipping.
Light processing is appropriate for: well-recorded content that needs minimal adjustment, acoustic music where dynamics are intentional, and content where natural dynamics should be preserved.
Moderate Standard broadcast compression suitable for most content. TNT analyzes your audio's average RMS level and applies moderate compression with a 3.5:1 ratio. The software calculates makeup gain automatically based on how much compression is being applied, ensuring consistent output levels without manual adjustment.
Moderate processing works for: podcasts, voice-overs, most music content, and general broadcast material that needs to sound consistent across different playback systems.
Broadcast Aggressive multiband processing for maximum clarity and consistency. Instead of analyzing overall dynamics, the software splits your audio into five frequency bands (sub-bass, bass, low-mid, mid, and high) and analyzes each band independently. Each band receives compression tailored to its specific characteristics—bass frequencies get tighter control with longer attack times, while high frequencies receive faster compression to maintain clarity.
The Broadcast preset uses adaptive ratios: bass content receives moderate compression (4.0:1 ratio) while high-frequency content gets more aggressive processing (up to 8.0:1 ratio). This ensures your audio maintains punch in the low end while achieving maximum intelligibility in the speech range. Attack and release times are frequency-dependent, ranging from 200ms in the bass to 100ms in the highs.
Broadcast processing is designed for: radio content, streaming platforms with varied playback systems, content consumed on small speakers or mobile devices, and any situation where maximum loudness and clarity are required.
EQ processing analyzes your audio's frequency response across ten octave-spaced bands from 50Hz to 12.8kHz+. The software measures RMS level, peak level, and crest factor for each band, then compares these measurements against professional target curves. All EQ adjustments use an attenuation-focused philosophy—corrections are calculated, then halved before application, with a maximum adjustment of ±10 dB. This conservative approach maintains audio quality while achieving broadcast standards.
No equalization is applied.
Targets a pink noise curve, which naturally contains more energy in the bass frequencies (-3 dB per octave rise from 1kHz). The Flat preset attenuates frequencies that exceed this curve while leaving frequencies below the curve unchanged. This prevents excessive energy buildup in any frequency range while maintaining the natural tonal balance of your content.
Flat EQ is appropriate for: music content where the original mix should be respected, material that's already well-balanced, and situations where you want to prevent frequency buildup without imposing a specific tonal character.
Optimized for vocal clarity and intelligibility. The Speech preset boosts presence frequencies (1.6kHz-6.4kHz) where consonants and speech intelligibility live, while attenuating sub-bass rumble and reducing boxiness in the 200-400Hz range. The target curve incorporates Fletcher-Munson loudness compensation to ensure speech remains clear at any listening level.
Specific adjustments in Speech mode:
Speech EQ is designed for: podcasts, voice-overs, audiobooks, conference recordings, interviews, and any content where vocal clarity is paramount.
Aggressive clarity enhancement for playback on small speakers, mobile devices, and varied listening environments. The Broadcast curve emphasizes midrange intelligibility even more strongly than Speech mode, with deeper cuts in the bass and more aggressive presence boost. This ensures your content cuts through on phone speakers, laptop audio, and car radios.
Specific adjustments in Broadcast mode:
Broadcast EQ is intended for: radio content, streaming platforms, mobile-first content, situations where playback systems are unknown, and any content that must remain intelligible on poor speakers.
When enabled, this checkbox disables both Dynamics and EQ processing regardless of their selected settings. Use this when you want loudness normalization only, without any dynamics control or tonal shaping. The Bypass option is useful for: testing how your audio sounds with normalization alone, A/B comparing processed versus unprocessed versions, or situations where you've already applied processing in your DAW and only need format conversion and loudness compliance.
When multiple processing stages are enabled, TNT applies them in this order:
This signal chain ensures frequency balance is corrected before dynamics processing, preventing the compressor from reacting to frequency imbalances. The de-esser removes harsh sibilance after EQ boosts but before compression, ensuring the compressor doesn't overreact to "s" sounds. Loudness normalization happens last, after all processing is complete, guaranteeing your target LUFS level is achieved accurately.
All processing happens at 192kHz sample rate internally to ensure intersample peak accuracy. For 16-bit PCM output, the software applies triangular dithering after all processing to minimize quantization artifacts. Multiband processing uses linear-phase crossover filters to prevent phase distortion between frequency bands.
The adaptive nature of TNT's processing means two identical preset selections may produce different filter parameters depending on the input audio's characteristics. This is intentional—the software adjusts its processing based on what it measures, ensuring optimal results for each file rather than applying static presets that may not suit the content.
Please note that these options are available to some users only after 5th December 2025.