I cannot find much information on sphinx training speed. There is only this
link: http://www.speech.cs.cmu.edu/sphinxman/FAQ.html#3
Are there any measurements on the training speed given the amount of
recordings, dictionary size, whether its Context Independent or context
dependent?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The speed depends only on database size and on number of CPUs.
There are no measurements for all possible combinations of processors and
database sizes. If you have some specific values in mind, you can ask about
them.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
we are using Amazon cloud with this configuration:
7.5 GB memory
4 EC2 Compute Units (2 virtual cores with 2 EC2 Compute Units each)
where each EC2 Compute Unit provides the equivalent CPU capacity of a 1.0-1.2
GHz 2007 Opteron or 2007 Xeon processor.
We have around 10hours of 5-10 seconds recordings.
Thanks for your help!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I cannot find much information on sphinx training speed. There is only this
link:
http://www.speech.cs.cmu.edu/sphinxman/FAQ.html#3
Are there any measurements on the training speed given the amount of
recordings, dictionary size, whether its Context Independent or context
dependent?
The speed depends only on database size and on number of CPUs.
There are no measurements for all possible combinations of processors and
database sizes. If you have some specific values in mind, you can ask about
them.
thanks,
we are using Amazon cloud with this configuration:
7.5 GB memory
4 EC2 Compute Units (2 virtual cores with 2 EC2 Compute Units each)
where each EC2 Compute Unit provides the equivalent CPU capacity of a 1.0-1.2
GHz 2007 Opteron or 2007 Xeon processor.
We have around 10hours of 5-10 seconds recordings.
Thanks for your help!
10 hours database is trained in less then an hour on single core.