I have a speech project that requires acoustic training to be done in code. I can successfully create training files with transcripts and related registry entries in Windows 7 using SAPI. However, I cannot determine if the Recognition Engine uses these files successfully and adapts its model. My questions are as follows:
When performing training through the control panel training user interface, the system stores training files in "{AppData] \ Local \ Microsoft \ Speech \ Files \ TrainingAudio". Should the audio device files be stored in this place or can I store them in another place if the registry entries for the profile reflect the correct path?
- The speech control panel creates registry entries for training audio files in the key "HKCU \ Software \ Microsoft \ Speech \ RecoProfiles \ Tokens {ProfileGUID} {00000000-0000-0000-0000-0000000000000000} \ Files".
a) Do you write the registry entries created by my training code HAVE in "{00000000-0000-0000-0000-0000000000000000} \ Files" or can I create a new random GUID in the {ProfileGUID} section?
b) Is the subsection connected to the Files file?
c) And follow the HAVE registry values ββto follow the form "TrainingAudio-xxxx-xxxxxxxx-xxxxxxxx" or can I use other values?
d) Finally, these registry values ββare of the form "% 1c% \ Microsoft \ Speech \ Files \ TrainingAudio \ SP-xxx .... xxx". Can I indicate an absolute path?
e) Should the HAVE file names follow the form "SP-xxx .... xxx.wav" or can I use any unique file names?
Thanks.
Kettlebell
source share