I've come back (again) to look at sphinx for a telephony asr. I asked a projects developer why they (FreeSwitch) bundled pocketsphinx instead of sphinx 4 with their software. One reason given was that PocketSphinx is "thread safe." I know that Sphinx 4 was written in Java so I'm curious to know whether it to is "thread safe" since the implication was it wasn't.
Also, reading over information about Sphinx 4, seems to me to suggest that it has more or better "things" in it like RASTA, support for multiple recognizers, speaker independent abilities, etc than PocketSphinx. Is this correct or better yet, what are the current advantages of one over the other?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've come back (again) to look at sphinx for a telephony asr. I asked a projects developer why they (FreeSwitch) bundled pocketsphinx instead of sphinx 4 with their software. One reason given was that PocketSphinx is "thread safe." I know that Sphinx 4 was written in Java so I'm curious to know whether it to is "thread safe" since the implication was it wasn't.
Also, reading over information about Sphinx 4, seems to me to suggest that it has more or better "things" in it like RASTA, support for multiple recognizers, speaker independent abilities, etc than PocketSphinx. Is this correct or better yet, what are the current advantages of one over the other?
> I know that Sphinx 4 was written in Java so I'm curious to know whether it to is "thread safe" since the implication was it wasn't.
It's thread safe as well. I suppose the issue is that sphinx4 is in Java, which is not that easy to reuse from C code.
> it has more or better "things" in it like RASTA, support for multiple recognizers, speaker independent abilities, etc than PocketSphinx.
no
> Is this correct or better yet, what are the current advantages of one over the other?
more flexible architecture because of high-level language used.