I had what is probably a stupid question about the CMU Sphinx II system. I want to use it to identify sound files that are 8bit, 8khz instead of 16bit 16 khz. Can this be done? I seem to remember reading that it could be trained with 8 bit sound files, so is that what I have to do, train it with 8bit, 8kHz files to make it identify them? Any help you can give me is greatly appreciated.
Thanks very much in advance,
-Andrew
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
You have two solutions:
- use 8 bit models (they exist for telephony)
- transform your 8 bits files in 16 bits by inserting zeroes between samples. Or by using a sound editor like cooledit96. You'll have slightly noisy 16 files (quantization noise).
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello all,
I had what is probably a stupid question about the CMU Sphinx II system. I want to use it to identify sound files that are 8bit, 8khz instead of 16bit 16 khz. Can this be done? I seem to remember reading that it could be trained with 8 bit sound files, so is that what I have to do, train it with 8bit, 8kHz files to make it identify them? Any help you can give me is greatly appreciated.
Thanks very much in advance,
-Andrew
Hi,
You have two solutions:
- use 8 bit models (they exist for telephony)
- transform your 8 bits files in 16 bits by inserting zeroes between samples. Or by using a sound editor like cooledit96. You'll have slightly noisy 16 files (quantization noise).