Reply To: RapidEars 2.04 – crash when using within Cocos2d

Home Forums OpenEars plugins RapidEars 2.04 – crash when using within Cocos2d Reply To: RapidEars 2.04 – crash when using within Cocos2d

#1026488
anhtu
Participant

Here is the contents of the files which you need. In this case , I tested from A-G. When came to letter C, it crashed.

lession_vocal_C_2.dic file

C S IY

lession_vocal_C_2.arpa

#############################################################################
## Copyright (c) 1996, Carnegie Mellon University, Cambridge University,
## Ronald Rosenfeld and Philip Clarkson
## Version 3, Copyright (c) 2006, Carnegie Mellon University 
## Contributors includes Wen Xu, Ananlada Chotimongkol, 
## David Huggins-Daines, Arthur Chan and Alan Black 
#############################################################################
=============================================================================
===============  This file was produced by the CMU-Cambridge  ===============
===============     Statistical Language Modeling Toolkit     ===============
=============================================================================
This is a 3-gram language model, based on a vocabulary of 3 words,
  which begins "</s>", "<s>", "C"...
This is a CLOSED-vocabulary model
  (OOVs eliminated from training data and are forbidden in test data)
Witten Bell discounting was applied.
This file is in the ARPA-standard format introduced by Doug Paul.

p(wd3|wd1,wd2)= if(trigram exists)           p_3(wd1,wd2,wd3)
                else if(bigram w1,w2 exists) bo_wt_2(w1,w2)*p(wd3|wd2)
                else                         p(wd3|w2)

p(wd2|wd1)= if(bigram exists) p_2(wd1,wd2)
            else              bo_wt_1(wd1)*p_1(wd2)

All probs and back-off weights (bo_wt) are given in log10 form.

Data formats:

Beginning of data mark: \data\
ngram 1=nr            # number of 1-grams
ngram 2=nr            # number of 2-grams
ngram 3=nr            # number of 3-grams

\1-grams:
p_1     wd_1 bo_wt_1
\2-grams:
p_2     wd_1 wd_2 bo_wt_2
\3-grams:
p_3     wd_1 wd_2 wd_3 

end of data mark: \end\

\data\
ngram 1=3
ngram 2=2
ngram 3=1

\1-grams:
-98.6990 </s>	0.0000
-98.6990 <s>	-99.9990
0.0000 C	-0.3010

\2-grams:
-0.3010 <s> C 0.0000
-0.3010 C </s> -0.3010

\3-grams:
-0.3010 <s> C </s> 

\end\

the log console

2015-08-04 16:37:00.758 App_Name[50514:4263641] Starting OpenEars logging for OpenEars version 2.03 on 64-bit device (or build): iPad running iOS version: 8.300000

2015-08-04 16:37:00.768 App_Name[50514:4263641] Starting dynamic language model generation

2015-08-04 16:37:00.892 App_Name[50514:4263641] Done creating language model with CMUCLMTK in 0.123174 seconds.

2015-08-04 16:37:00.900 App_Name[50514:4263641] I'm done running performDictionaryLookup and it took 0.002335 seconds

2015-08-04 16:37:00.908 App_Name[50514:4263641] I'm done running dynamic language model generation and it took 0.148800 seconds

2015-08-04 16:37:00.908 App_Name[50514:4263641] Valid setSecondsOfSilence value of 0.200000 will be used.

2015-08-04 16:37:01.017 App_Name[50514:4264050] there is a request to change to the language model file /var/mobile/Containers/Data/Application/A9D01313-E149-4F27-9067-EB2DA15AB639/Library/Caches/lession_vocal_B_1.DMP

2015-08-04 16:37:01.017 App_Name[50514:4264050] The language model ID is 1438681021

2015-08-04 16:37:01.032 App_Name[50514:4264050] Success loading the specified dictionary file /var/mobile/Containers/Data/Application/A9D01313-E149-4F27-9067-EB2DA15AB639/Library/Caches/lession_vocal_B_1.dic.

2015-08-04 16:37:01.033 App_Name[50514:4264050] Success loading the specified language model file /var/mobile/Containers/Data/Application/A9D01313-E149-4F27-9067-EB2DA15AB639/Library/Caches/lession_vocal_B_1.DMP.

2015-08-04 16:37:01.033 App_Name[50514:4263641] Pocketsphinx is now using the following language model: 

/var/mobile/Containers/Data/Application/A9D01313-E149-4F27-9067-EB2DA15AB639/Library/Caches/lession_vocal_B_1.DMP and the following dictionary: /var/mobile/Containers/Data/Application/A9D01313-E149-4F27-9067-EB2DA15AB639/Library/Caches/lession_vocal_B_1.dic

2015-08-04 16:37:01.039 App_Name[50514:4264050] Changed language model. Project has these words or phrases in its dictionary:

B

2015-08-04 16:37:01.392 App_Name[50514:4264050] Speech detected...

2015-08-04 16:37:01.393 App_Name[50514:4264050] Pocketsphinx heard "" with a score of (-2680) and an utterance ID of 3.

2015-08-04 16:37:01.394 App_Name[50514:4264050] Hypothesis was null so we aren't returning it. If you want null hypotheses to also be returned, set OEPocketsphinxController's property returnNullHypotheses to TRUE before starting OEPocketsphinxController.

2015-08-04 16:37:01.544 App_Name[50514:4264050] Pocketsphinx heard "B" with a score of (-5403) and an utterance ID of 4.

2015-08-04 16:37:01.585 App_Name[50514:4263641] rapidEarsDidReceiveFinishedSpeechHypothesis: B with score: -5403

2015-08-04 16:37:01.775 App_Name[50514:4263763] End of speech detected...

2015-08-04 16:37:01.779 App_Name[50514:4263641] pocketsphinxDidDetectFinishedSpeech

2015-08-04 16:37:01.785 App_Name[50514:4263763] Pocketsphinx heard "B" with a score of (-8131) and an utterance ID of 5.

2015-08-04 16:37:01.792 App_Name[50514:4263641] rapidEarsDidReceiveFinishedSpeechHypothesis: B with score: -8131

2015-08-04 16:37:03.404 App_Name[50514:4263641] Starting OpenEars logging for OpenEars version 2.03 on 64-bit device (or build): iPad running iOS version: 8.300000

2015-08-04 16:37:03.416 App_Name[50514:4263641] Starting dynamic language model generation

2015-08-04 16:37:03.517 App_Name[50514:4263641] Done creating language model with CMUCLMTK in 0.100489 seconds.

2015-08-04 16:37:03.528 App_Name[50514:4263641] I'm done running performDictionaryLookup and it took 0.004014 seconds

2015-08-04 16:37:03.536 App_Name[50514:4263641] I'm done running dynamic language model generation and it took 0.131257 seconds

2015-08-04 16:37:03.536 App_Name[50514:4263641] Valid setSecondsOfSilence value of 0.200000 will be used.

2015-08-04 16:37:03.567 App_Name[50514:4264074] there is a request to change to the language model file /var/mobile/Containers/Data/Application/A9D01313-E149-4F27-9067-EB2DA15AB639/Library/Caches/lession_vocal_C_2.DMP

2015-08-04 16:37:03.569 App_Name[50514:4264074] The language model ID is 1438681023

App_Name(50514,0x104f10000) malloc: *** error for object 0x900000008: pointer being freed was not allocated

*** set a breakpoint in malloc_error_break to debug

call stack

Thread 16Queue : com.apple.root.default-qos (concurent)
#0	0x00000001948c7270 in __pthread_kill ()
#1	0x0000000194965170 in pthread_kill ()
#2	0x000000019483eb18 in abort ()
#3	0x00000001948f42fc in free ()
#4	0x0000000100473054 in ngram_model_set_map_words ()
#5	0x0000000100476598 in ___lldb_unnamed_function156$$App_Name ()
#6	0x00000001004693e4 in ps_load_dict ()
#7	0x0000000100479c10 in ___lldb_unnamed_function183$$App_Name ()
#8	0x00000001004611c8 in ___lldb_unnamed_function78$$App_Name ()
#9	0x0000000100461858 in ___lldb_unnamed_function82$$App_Name ()
#10	0x0000000101388f94 in _dispatch_client_callout ()
#11	0x00000001013a0848 in _dispatch_source_latch_and_call ()
#12	0x000000010138b1c0 in _dispatch_source_invoke ()
#13	0x00000001013965d4 in _dispatch_root_queue_drain ()
#14	0x0000000101398248 in _dispatch_worker_thread3 ()
#15	0x000000019496122c in _pthread_wqthread ()

P.S: when I use REJECTO (demo version), this problem doesn’t appear. Here is the code I use:

 error = [_languageModelGenerator generateRejectingLanguageModelFromArray:arrWords
                                                            withFilesNamed:name
                                                    withOptionalExclusions:[GlobalData sharedInstance].optionalExclusion
                                                           usingVowelsOnly:[GlobalData sharedInstance].usingVowelsOnly
                                                                withWeight:[NSNumber numberWithFloat: [GlobalData sharedInstance].weight_rejecto]
                                                    forAcousticModelAtPath:[OEAcousticModel pathToModel:@"AcousticModelEnglish"]];