Is there a program that can convert words to their respective phonemes? It doesn't really matter which syntax/format is used as long as its consistent and reproducible. I have lots of data to convert but CMU's text to Arpabet converter only converted about 100 of them. I tried running "espeak [word] -x" to get espeak's code, but I'd have to wait for the program to say the word before printing. Also, I can't install espeak on the cluster that I'm running my data on.
Asked
Active
Viewed 1,533 times
0
-
What kind of words are you trying to convert? – rmalouf Jun 15 '16 at 20:38
-
I have a list of words from a online chat forum and I'm hoping it will help handle misspellings. – Jonathan Jun 15 '16 at 20:57
-
1check this out: http://www.nltk.org/_modules/nltk/corpus/reader/cmudict.html – alvas Jun 15 '16 at 23:49
-
Have you tried normalizing the spellings with a spellchecker such as the Open Office one or [PyEnchant](http://pythonhosted.org/pyenchant/tutorial.html)? I don't really know how a phonetic transcription a la CMUDict really helps here... – patrick Jun 16 '16 at 15:34