| |
Speech Synthesis
CSRE’s speech synthesizer provides a full implementation of Klatt parametric speech
synthesizer, providing precise control over 40 specific speech parameters. Parameters
include the frequencies, bandwidths, and amplitudes of individual formants, and
the type and amplitude of the speech sound source as a function of time. Both
parallel and cascade synthesis is permitted.
Much attention has been given to the design of features to facilitate synthesis.
The system is controlled by an easy-to-use, mouse-based interface. All Klatt synthesis
parameters are visible in scrolling pick lists.
Default values are provided for all parameters so that only specific, desired
parameters need be specified. Our “knot editor” facility automatically interpolates
between specified parameter values so that you need only specify when and how
parameter values change. Parameters are checked automatically, preventing synthesis
errors due to out-of-range parameters. Signal preview allows immediate playback
and display of synthesized signals.
CSRE's multi-dimensional continuum generator facilitates the synthesis of speech
continua formed through systematic variation in one or more parameters. Signals
synthesized in this manner are labelled and stored automatically, in a systematic
manner. The synthesizer reads data files generated by CSRE’s formant and pitch
analysis modules, permitting high-quality “copy-synthesis” and greatly facilitating
the synthesis of new speech signals.
Order CSRE | Download
CSRE Demo
|
|
