Time Tick, ms - this is a minimal amount of time, considered by the recognition algorithm. Every note duration is multiply of Time Tick and every note starting time is multiply of Time Tick.
Decreasing this parameter always increases memory requirements of the recognition process, and in the case of FFT it also increases the computational time as well. Recommended range for this parameter is 7..40 ms.
This method is relatively fast. It uses a sophisticated auto tuning and harmonic filter algorithm. This allows you to get very little amount of refuse notes - compare spectral diagrams made by Sensors and by WIDI 2.5 method. Nevertheless it has its own disadvantage - weak notes may be lost on the very first step of recognition. When having the main voice much more louder the than accompaniment, and you need to get only main voice - it's a good idea to use this method.
Auto Tuning - auto tuning of Sensors to exact components that exist in the source. It is useful when audio file you are transcribing is recorded with tones, shifted relatively standard frequencies.
Harmonic Filter - it reduces the amount of overtones that are recognized as separate notes.
Monophonic Mode detects only the loudest voice from a source file. It is recommended to set this parameter when recognizing one-voice melodies.
Time Shift of Bass Notes corrects starting times of bass notes compared to trebles. Sensors usually pushes bass notes further in time, so increasing of this parameter shifts them back. So if you hear that bass voices are sounding after trebles, try to change this parameter. Reasonable values are usually 0..40% of the scale.
This is a new algorithm that is created for recognition of monophonic melodies only. It uses completely different technique than Sensors.
Auto Tuning performs tuning to exact components that exist in the source. It is useful when an audio file you are transcribing is recorded with tones, shifted relatively standard frequencies (for instance, human voice without the accompaniment).
It is an algorithm used since WIDI 2.5. Usually it produces more notes than Advanced Sensors, so use it when Advanced Sensors loses notes.
Time Shift of Bass Notes - corrects starting times of bass notes compared to trebles. Sensors usually pushes bass notes further in time, so increasing of this parameter shifts them back. So if you hear that bass voices are sounding after trebles, try to change this parameter. Reasonable values are usually 0..40% of the scale.
This is an older algorithm, based on Fast Fourier Transformation. In general, Sensors is almost always better than FFT, except the case when the music is played by incorrectly tuned instruments (for example, if note C lies a quarter tone higher than standard C).
Time Shift of Bass Notes - corrects starting times of bass notes compared to trebles. FFT brings the bass notes before the time they should sound at, and increasing the "Time Shift" shifts them further. So if you hear that bass voices are sounding before trebles, try to change this parameter. Reasonable values are usually 0..70% of the scale.
|
