[neshe9]

Hi,
I was able to reproduce the same scenario..
Please find the log below:

2015-01-06 10:07:48.634 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx is now listening.
2015-01-06 10:07:48.638 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx started.
2015-01-06 10:07:49.116 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected speech.
2015-01-06 10:07:52.392 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
2015-01-06 10:07:52.490 OpenEarsSampleApp[194:3912] Local callback: The received hypothesis is START with a score of -1208 and an ID of 0
2015-01-06 10:07:52.670 OpenEarsSampleApp[194:3912] Local callback: Flite has started speaking
2015-01-06 10:07:52.675 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has suspended recognition.
2015-01-06 10:07:54.717 OpenEarsSampleApp[194:3912] Local callback: Flite has finished speaking
2015-01-06 10:07:54.723 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has resumed recognition.
2015-01-06 10:07:55.211 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected speech.
2015-01-06 10:08:01.346 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
2015-01-06 10:08:01.459 OpenEarsSampleApp[194:3912] Local callback: The received hypothesis is CHANGE MODEL with a score of -17461 and an ID of 1
2015-01-06 10:08:01.678 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx is now using the following language model:
/var/mobile/Containers/Data/Application/0AF6B9CF-9B3B-4D17-AD08-FC2EF042CFE1/Library/Caches/SecondOpenEarsDynamicLanguageModel.DMP and the following dictionary: /var/mobile/Containers/Data/Application/0AF6B9CF-9B3B-4D17-AD08-FC2EF042CFE1/Library/Caches/SecondOpenEarsDynamicLanguageModel.dic
2015-01-06 10:08:01.678 OpenEarsSampleApp[194:3912] Local callback: Flite has started speaking
2015-01-06 10:08:01.683 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has suspended recognition.
2015-01-06 10:08:01.736 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected speech.
2015-01-06 10:08:04.190 OpenEarsSampleApp[194:3912] Local callback: Flite has finished speaking
2015-01-06 10:08:04.195 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has resumed recognition.
2015-01-06 10:08:04.421 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected speech.
2015-01-06 10:08:17.086 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
2015-01-06 10:08:47.247 OpenEarsSampleApp[194:3912] Local callback: The received hypothesis is TO TO CAN YOU PLEASE REPEAT with a score of -14960 and an ID of 2
2015-01-06 10:08:47.469 OpenEarsSampleApp[194:3912] Local callback: Flite has started speaking
2015-01-06 10:08:47.474 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has suspended recognition.
2015-01-06 10:08:50.353 OpenEarsSampleApp[194:3912] Local callback: Flite has finished speaking
2015-01-06 10:08:50.359 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has resumed recognition.
2015-01-06 10:08:50.862 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected speech.
2015-01-06 10:09:21.102 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
2015-01-06 10:11:19.424 OpenEarsSampleApp[194:3912] Received memory warning. — HERE MEMORY USAGE WAS AROUND 550 MB
2015-01-06 10:12:43.160 OpenEarsSampleApp[194:3912] Local callback: The received hypothesis is GO TO NEXT SLIDE with a score of -12202 and an ID of 3
2015-01-06 10:12:43.394 OpenEarsSampleApp[194:3912] Local callback: Flite has started speaking
2015-01-06 10:12:43.400 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has suspended recognition.
2015-01-06 10:12:46.079 OpenEarsSampleApp[194:3912] Local callback: Flite has finished speaking
2015-01-06 10:12:46.084 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has resumed recognition.
2015-01-06 10:12:48.519 OpenEarsSampleApp[194:3912] Local callback: Pocketsphinx has detected speech.

I will get through the recording and vocabulary to you soon.

[neshe9]

Sorry for the confusion..
Yes, initially it was about null hypothesis. That is solved now.
Later, it is the issue of slow recognition. Sometimes it takes as long as even 2-3 minutes to return a hypothesis.(Which leads to memory warning followed by the app crashing).

So I replaced my language array in the sample app and tried. Recognition is slow there too.. So I asked if it was a matter of the size of the vocabulary..

As you suggested, I will try varying the vadThreshold and the secondsOfSilence to see any changes in the performance.
Also, I have removed the ‘suspendRecognition’ call now.
Thank you. I will keep you posted.

[neshe9]

In the language array, I have around 20 strings including some phrases and small sentences.
When I was using the previous version, there was no problems with recognition.
Is this a limitation of the version 2.0 to recognize only words?

[neshe9]

Now I see that I was suspending recognition in the ‘pocketsphinxDidCompleteCalibration’ method.
Is there any method similar to that in v2.0?
I didn’t find any methods replacing the above mentioned..

[Author]

Posts