User Guide

Chapter 15 EVOC 20 PolySynth 225
Editing the Analysis and Synthesis Signals
The following section outlines how you can edit the analysis and synthesis signals to
achieve better speech intelligibility.
Compressing the Analysis Signal
The less the level changes, the better the intelligibility of the vocoder. You should
therefore compress the analysis signal in most cases.
Enhancing High Frequency Energy
The vocoder, in a way, always generates the intersection point of the analysis and
synthesis signals. To explain: If there’s no treble portion in the analysis signal, the
resulting vocoder output will also lack treble. This is also the case when the synthesis
signal features a lot of high frequency content. This is true of each frequency band. As
such, the vocoder demands a stable level in all frequency bands from both input
signals, in order to obtain the best results.
Due to the way human beings hear, the intelligibility of speech is highly dependent on
the presence of high frequency content. To aid in keeping speech clear, it may be
worthwhile using equalization to boost or cut particular frequencies in analysis signals
before processing them with the vocoders.
If the analysis signal consists of vocals or speech, a simple shelving filter should be
sufficient. It doesn’t require much processing power, and efficiently boosts the high-
mid and treble range, which is so important for speech intelligibility.
If the synthesis signal lacks treble energy, it can be generated with a distortion effect.
The Overdrive effect is perfect for this purpose (see “Overdrive” on page 56).
Avoiding Sonic Artifacts
A common problem with vocoder sounds are sudden signal interruptions (ripping,
breaking sounds) and rapidly triggered noises, during speech pauses.
Release Parameter in the Analysis Section
The Release parameter defines the speed that a given synthesis frequency band can
decrease in level, if the signal level of the respective analysis band decreases abruptly.
The sound is smoother when the band levels decrease slowly. To achieve this smoother
character, use higher Release values in the analysis section of the interface. Longer
release times result in a washy sound.
Short Attack values are no problem. They may, in fact, even be desirable when a fast
reaction to impulse signals by the vocoder is required.