Trying to do silence detection on Androd device using processRaw.
I have the following config setup:
en-us.lm.bin language model
en-us-ptm acoustic model
cmudict-en-us.dict dictionary
also setting remove_noise to True and samprate to 8000
also setKeywordThreshold(1e-1f) but also tried other values (1e-1f, 1e-5f, 1e-10f, 1e-15f, 1e-20f, 1e-25f, 1e-30f, 1e-35f, 1e-40f)
Setting for Ngram Search.
Results are poor, I get around 60% good silence detetion and in the other times I get false detection. Need to mention that in general the word detection is wrong (100% wrong).
What is missing?
Thanks.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
Trying to do silence detection on Androd device using processRaw.
I have the following config setup:
en-us.lm.bin language model
en-us-ptm acoustic model
cmudict-en-us.dict dictionary
also setting remove_noise to True and samprate to 8000
also setKeywordThreshold(1e-1f) but also tried other values (1e-1f, 1e-5f, 1e-10f, 1e-15f, 1e-20f, 1e-25f, 1e-30f, 1e-35f, 1e-40f)
Setting for Ngram Search.
Results are poor, I get around 60% good silence detetion and in the other times I get false detection. Need to mention that in general the word detection is wrong (100% wrong).
What is missing?
Thanks.