Podcaster, blogger — recording processing
Musician — demo preparation, mastering
Production studio — routine processing of large audio volumes
Educator — cleaning up lecture recordings
Anyone working with audio — who doesn't want to learn a DAW
Full audio processing: noise reduction, normalization, compression, EQ, trimming — without a sound engineer
You can just say: "Clean and normalize" — GistiQ will choose the parameters automatically
Sox two-pass — creates a noise profile from a quiet segment, then suppresses across the entire file. Best result for stationary noise (AC, fan, hum). RNNoise (FFmpeg) — neural noise suppression in real time. For complex noise (street, cafe). FFT-based (afftdn) — frequency-domain noise suppression for fine tuning. Noise gate — automatic suppression of sound below threshold (quiet background noise between phrases).
EBU R128 (loudnorm) — YouTube, Spotify, Apple Podcasts standard. All files equally loud. Compressor — quiet parts louder, loud parts quieter. Comfortable listening without volume adjustment. Limiter — preventing clipping (overload). Peak normalization — bringing to a specified level.
Parametric equalizer — 18 bands, precise frequency tuning. Highpass — remove hum, low-frequency noise (below 80 Hz). Lowpass — remove hiss, high-frequency noise. Bass/Treble — quick low and high correction. De-essing — reducing sibilant sounds.
Silence removal — from beginning, end, and middle of recording with adjustable threshold. Silence detection — automatic pause detection for splitting. VAD (Voice Activity Detection) — identifying segments with voice.
Joining multiple audio files with crossfade. Music overlay under voice with automatic ducking (lowering music when speaker talks). Fade in/out. Tempo change without pitch change (and vice versa). Mono ↔ stereo conversion.
Any format → any format: WAV, MP3, FLAC, OGG, M4A, AAC, AIFF. Bitrate, sample rate, channel configuration. Spectrogram — frequency visualization as PNG (for quality analysis).
A professor records lectures on a phone voice recorder. 20 recordings, 40–90 minutes each. Quality — auditorium noise, door slams, uneven volume, long pauses.
GistiQ batch processes all 20 files: Sox noise reduction (profile from quiet segment) → highpass 80 Hz (remove hum) → remove pauses longer than 3 seconds → dynamic range compression → EBU R128 normalization → convert to MP3 128 kbps. Output — 20 clean, even recordings ready for listening or transcription. Before/after spectrograms show the difference.