I am also exploring Praat, htk tool kits? Is sphinx efficient than these two tool kits?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2008-12-01
The only problem I have read about Sphinx and large training sources is that Sphinx is slow for extremely large audio files. Rather than feeding it an entire novel (or even a whole paragraph), for example, break the audio source into sentences.
Something tells me you'll just be feeding it individual words, though. It seems to be pretty snappy for me.
Sphinx does continuous speech.
I have no idea how long it would take to train 100,000 words. I would suggest trying 100 or 1,000 words as a starting sample. That would give you an idea for the total duration.
Everything I have read says that Sphinx is the best open source ASR engine available. I believe it.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear Sir,
Please give me approximate time estimate?
I am also exploring Praat, htk tool kits? Is sphinx efficient than these two tool kits?
The only problem I have read about Sphinx and large training sources is that Sphinx is slow for extremely large audio files. Rather than feeding it an entire novel (or even a whole paragraph), for example, break the audio source into sentences.
Something tells me you'll just be feeding it individual words, though. It seems to be pretty snappy for me.
Sphinx does continuous speech.
I have no idea how long it would take to train 100,000 words. I would suggest trying 100 or 1,000 words as a starting sample. That would give you an idea for the total duration.
Everything I have read says that Sphinx is the best open source ASR engine available. I believe it.