[Speex-dev] Use voice onset timing to identify voiceless

[Matt Robinson]

In increasing the computation time and bit rate with VBR:

Has anyone considered implementing the standard audiological
recognition technique of using the duration of zero energy (voice
onset timing) to identify the presence of voiceless sounds?

I would like a means of determining whether or not a given window will
be full of voiced speech or not.

Matt
--
The swallow may fly south with the sun or the house martin or the
plover may seek warmer climes in winter yet these are not strangers to
our land.