I`ve already achieved training prosess with AN4_clstk, now I begin decoding with an4test_clstk, but it can not go well in the comman window it displayed just 0%,sentence error 49.231%, 7errors
when I Look up into logdir/decode,I found the follwing errors
INFO: f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c(1413): Trying to read F:/train/an4/etc/an4.ug.lm as precompiled dump file
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1425: Wrong magic header size number 23232323: F:/train/an4/etc/an4.ug.lm is not a dump file
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: '<UNK>' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'HALL' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'LANE' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'MEMORY' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'TWELVTH' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'WEAN' is in LM unigrams but not in dictionary
I want to know what happened. I am looking forward to your reply,Thank you !
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
> ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1425: Wrong magic header size number 23232323: F:/train/an4/etc/an4.ug.lm is not a dump file
It's not a critical error, you can just ignore it.
> ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'WEAN' is in LM unigrams but not in dictionary
This one too. As you could understand from a error message the word is missing in the dictionary. To fix this situation you must open the dictionary with a text editor and add a word with a transcription there. Remember that words must be sorted. The dictionary is located at etc/an4.dic
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
hello
decoder:pocketsphinx
I`ve already achieved training prosess with AN4_clstk, now I begin decoding with an4test_clstk, but it can not go well in the comman window it displayed just 0%,sentence error 49.231%, 7errors
when I Look up into logdir/decode,I found the follwing errors
INFO: f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c(1413): Trying to read F:/train/an4/etc/an4.ug.lm as precompiled dump file
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1425: Wrong magic header size number 23232323: F:/train/an4/etc/an4.ug.lm is not a dump file
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: '<UNK>' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'HALL' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'LANE' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'MEMORY' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'TWELVTH' is in LM unigrams but not in dictionary
ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'WEAN' is in LM unigrams but not in dictionary
I want to know what happened. I am looking forward to your reply,Thank you !
> sentence error 49.231%
It's the expected result
> ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1425: Wrong magic header size number 23232323: F:/train/an4/etc/an4.ug.lm is not a dump file
It's not a critical error, you can just ignore it.
> ERROR: "f:\train\pocketsphinx\src\libpocketsphinx\lm_3g.c", line 1033: 'WEAN' is in LM unigrams but not in dictionary
This one too. As you could understand from a error message the word is missing in the dictionary. To fix this situation you must open the dictionary with a text editor and add a word with a transcription there. Remember that words must be sorted. The dictionary is located at etc/an4.dic