The adaptive filtration (AF) algorithm
The adaptive filtration (AF) algorithmsmartVES ensures the maximization of the STI coefficient and maintains the appropriate headroom (5, 10 or 15 dB) between the useful signals, e.g. spoken messages, and any undesirable acoustic background.
This process of the maximisation of the STI is calculated by a number of innovative algorithms implemented in the SMART-DU-1604 unit. The key algorithms include the algorithm for adaptive filtering and temporal transposition of the speech signal, as well as algorithms for calculating SNR, STI and auto-calibration of the system, supporting real-time measurements, whose task is to obtain information that allows optimal adjustment of the sound to the current acoustic conditions.
This process of the maximisation of the STI is calculated by a number of innovative algorithms implemented in the SMART-DU-1604 unit. The key algorithms include the algorithm for adaptive filtering and temporal transposition of the speech signal, as well as algorithms for calculating SNR, STI and auto-calibration of the system, supporting real-time measurements, whose task is to obtain information that allows optimal adjustment of the sound to the current acoustic conditions.
The speech temporal transposition algorithm (STTA)
The speech temporal transposition algorithm (STTA) naturally and evenly changes the pace and duration of messages spoken in real time through smartVES system microphones (DFMS, DMS, DMS-LCD). The algorithm can distinguish the type of voice (male / female), determines the rate of speech, and most importantly, detects and shortens the duration of stuttering, while stuttering is understood as excessive prolongation of the articulation of the selected sound.