Explicit speech context

Protects spectral balance during announcements explicitly marked as speech.

How it works

The lab engine accepts known speech intervals: Auto EQ and Bass EQ preserve their gains, and speech is excluded from analysis used to track the music spectrum. Filters keep processing and leveling remains active. Intervals can be controlled through the engine or API; the online detector and synchronized flow have been developed and tested separately. Complete speech integration with a real station microphone remains open.

Use and limits

Results depend on the selected profile, source material, input level and distribution chain. Listen at matched loudness and check output after the codec before keeping a preset.