Inside the Enhancement
The enhancement is our secret sauce, with processing steps tailored to each individual recording. Here's a summary.
Notch EQ
Recognizes constant, pitched noise sources (such as buzzing light dimmers) and reduces them significantly.
Crosstalk reduction
An AI-based method for reducing crosstalk between the voice tracks. Bleeding between mics is far less of a problem.
Active speaker detection
Intelligent analysis of your audio recordings to detect what microphone is in use and serves as an assistant to other enhancement features.
De-noising
A powerful, AI-based de-noiser that identifies and reduces background and handling noise.
Tone-shaping EQ
Shapes the frequency content of voices to make them sound more pleasant.
High pass filter
Greatly reduces frequencies below the range of the speaking voice, e.g. electrical hum.
Adaptive leveler
Levels out loudness differences in speech within a single track and between tracks without affecting the noise floor.
De-esser
Removes excessive sibilance in your recording when any hissing s-like sounds are identified.
Mute inactive speakers
AI-based detection of speaker activity in each voice track is used to reduce levels in inactive intervals.
Loudness normalization
Normalize the overall level of the mix to a standard target (e.g. broadcasting -23 LUFS).
Stereo vs spatial
Spread out voices spatially for improved headphone listening by leveraging localisation data from our Stellar Mics.