3 different Alexa modules, which one is the best?

borrigan · Apr 25, 2017, 11:15 PM

@romain thank you very very much for all this feedback. I was definitely missing some pieces of the puzzle. I will try this again today and post my findings here. Have a great day!

romain · Apr 26, 2017, 8:16 AM

@johnnyboy I struggle because of this @johnnyboy at first. But the lambda service is working both with us and ireland (west-eu or eu-west. don’t remember exactly now).
I end up have skill that was working. On the amazon interfaces anyway by choosing that instead of us. (even though I’m in france. I had no error anymore on the amazon test aera)
I had to change the hard coded region in the code though (easy fix).
I think I had no more error at the end (don’t remember for sure). But nothing was happening when I talked (maybe it was again a microphone issue. Not really sure about that)
I also did not understood how to use the npm start dev correctly at that time so maybe I would be able to do a better job at finding the issue now.
Maybe I try today, maybe I never try again. we’ll see (it’s not my raspberry pie I using so I am limited in when I can use it. So I rather do what I am ask to first before trying again)

johnnyboy · Apr 28, 2017, 5:34 PM

This post is deleted!

romain · Apr 26, 2017, 9:30 AM

@johnnyboy Yes, But it’s not your fault. I have read an entire very long thread about mirror mirror on the wall (and maybe some others, I don’t remember if all the information were on the same spot or not) and it seem that one guy talked about trying that (changing the region) and the creator of the module himself didn’t knew Ireland was possible. But he did point out in which file the region thing was. (it wasn’t had to find though).
So I do think he meant that “that one ‘must’ be in the us-east” because he didn’t knew there was another option.
And he didn’t changed his tutorial to reflect that. (I’m guessing because he wasn’t able to test this way itself. Or maybe he wasn’t willing to)

borrigan · Apr 27, 2017, 1:05 AM

@romain hi again!!

I have done some progress. I followed your instructions, created the asound.conf file and setup myTest as the input device in the config.yaml file. When I run alexaPi, now it does not give me a microphone error, but a strange “Capture data too large” error. I get to hear the “hello” sound but when I speak “alexa” it won’t do anything.

Really weird. Here is the full output of main.py execution:

pi@raspberrypi:/opt/AlexaPi/src $ /opt/AlexaPi/src/main.py -d
INFO: pocketsphinx.c(152): Parsed model-specific feature parameters from /usr/local/lib/python2.7/dist-packages/pocketsphinx/model/en-us/feat.params
Current configuration:
[NAME]			[DEFLT]		[VALUE]
-agc			none		none
-agcthresh		2.0		2.000000e+00
-allphone				
-allphone_ci		no		no
-alpha			0.97		9.700000e-01
-ascale			20.0		2.000000e+01
-aw			1		1
-backtrace		no		no
-beam			1e-48		1.000000e-48
-bestpath		yes		yes
-bestpathlw		9.5		9.500000e+00
-ceplen			13		13
-cmn			live		batch
-cmninit		40,3,-1		41.00,-5.29,-0.12,5.09,2.48,-4.07,-1.37,-1.78,-5.08,-2.05,-6.45,-1.42,1.17
-compallsen		no		no
-debug					0
-dict					/usr/local/lib/python2.7/dist-packages/pocketsphinx/model/cmudict-en-us.dict
-dictcase		no		no
-dither			no		no
-doublebw		no		no
-ds			1		1
-fdict					
-feat			1s_c_d_dd	1s_c_d_dd
-featparams				
-fillprob		1e-8		1.000000e-08
-frate			100		100
-fsg					
-fsgusealtpron		yes		yes
-fsgusefiller		yes		yes
-fwdflat		yes		yes
-fwdflatbeam		1e-64		1.000000e-64
-fwdflatefwid		4		4
-fwdflatlw		8.5		8.500000e+00
-fwdflatsfwin		25		25
-fwdflatwbeam		7e-29		7.000000e-29
-fwdtree		yes		yes
-hmm					/usr/local/lib/python2.7/dist-packages/pocketsphinx/model/en-us
-input_endian		little		little
-jsgf					
-keyphrase				alexa
-kws					
-kws_delay		10		10
-kws_plp		1e-1		1.000000e-01
-kws_threshold		1		1.000000e-10
-latsize		5000		5000
-lda					
-ldadim			0		0
-lifter			0		22
-lm					
-lmctl					
-lmname					
-logbase		1.0001		1.000100e+00
-logfn					
-logspec		no		no
-lowerf			133.33334	1.300000e+02
-lpbeam			1e-40		1.000000e-40
-lponlybeam		7e-29		7.000000e-29
-lw			6.5		6.500000e+00
-maxhmmpf		30000		30000
-maxwpf			-1		-1
-mdef					
-mean					
-mfclogdir				
-min_endfr		0		0
-mixw					
-mixwfloor		0.0000001	1.000000e-07
-mllr					
-mmap			yes		yes
-ncep			13		13
-nfft			512		512
-nfilt			40		25
-nwpen			1.0		1.000000e+00
-pbeam			1e-48		1.000000e-48
-pip			1.0		1.000000e+00
-pl_beam		1e-10		1.000000e-10
-pl_pbeam		1e-10		1.000000e-10
-pl_pip			1.0		1.000000e+00
-pl_weight		3.0		3.000000e+00
-pl_window		5		5
-rawlogdir				
-remove_dc		no		no
-remove_noise		yes		yes
-remove_silence		yes		yes
-round_filters		yes		yes
-samprate		16000		1.600000e+04
-seed			-1		-1
-sendump				
-senlogdir				
-senmgau				
-silprob		0.005		5.000000e-03
-smoothspec		no		no
-svspec					0-12/13-25/26-38
-tmat					
-tmatfloor		0.0001		1.000000e-04
-topn			4		4
-topn_beam		0		0
-toprule				
-transform		legacy		dct
-unit_area		yes		yes
-upperf			6855.4976	6.800000e+03
-uw			1.0		1.000000e+00
-vad_postspeech		50		50
-vad_prespeech		20		20
-vad_startspeech	10		10
-vad_threshold		2.0		2.000000e+00
-var					
-varfloor		0.0001		1.000000e-04
-varnorm		no		no
-verbose		no		no
-warp_params				
-warp_type		inverse_linear	inverse_linear
-wbeam			7e-29		7.000000e-29
-wip			0.65		6.500000e-01
-wlen			0.025625	2.562500e-02

INFO: feat.c(715): Initializing feature stream to type: '1s_c_d_dd', ceplen=13, CMN='batch', VARNORM='no', AGC='none'
INFO: acmod.c(166): Using subvector specification 0-12/13-25/26-38
INFO: mdef.c(518): Reading model definition: /usr/local/lib/python2.7/dist-packages/pocketsphinx/model/en-us/mdef
INFO: mdef.c(531): Found byte-order mark BMDF, assuming this is a binary mdef file
INFO: bin_mdef.c(336): Reading binary model definition: /usr/local/lib/python2.7/dist-packages/pocketsphinx/model/en-us/mdef
INFO: bin_mdef.c(516): 42 CI-phone, 137053 CD-phone, 3 emitstate/phone, 126 CI-sen, 5126 Sen, 29324 Sen-Seq
INFO: tmat.c(149): Reading HMM transition probability matrices: /usr/local/lib/python2.7/dist-packages/pocketsphinx/model/en-us/transition_matrices
INFO: acmod.c(117): Attempting to use PTM computation module
INFO: ms_gauden.c(127): Reading mixture gaussian parameter: /usr/local/lib/python2.7/dist-packages/pocketsphinx/model/en-us/means
INFO: ms_gauden.c(242): 42 codebook, 3 feature, size: 
INFO: ms_gauden.c(244):  128x13
INFO: ms_gauden.c(244):  128x13
INFO: ms_gauden.c(244):  128x13
INFO: ms_gauden.c(127): Reading mixture gaussian parameter: /usr/local/lib/python2.7/dist-packages/pocketsphinx/model/en-us/variances
INFO: ms_gauden.c(242): 42 codebook, 3 feature, size: 
INFO: ms_gauden.c(244):  128x13
INFO: ms_gauden.c(244):  128x13
INFO: ms_gauden.c(244):  128x13
INFO: ms_gauden.c(304): 222 variance values floored
INFO: ptm_mgau.c(476): Loading senones from dump file /usr/local/lib/python2.7/dist-packages/pocketsphinx/model/en-us/sendump
INFO: ptm_mgau.c(500): BEGIN FILE FORMAT DESCRIPTION
INFO: ptm_mgau.c(563): Rows: 128, Columns: 5126
INFO: ptm_mgau.c(595): Using memory-mapped I/O for senones
INFO: ptm_mgau.c(838): Maximum top-N: 4
INFO: phone_loop_search.c(114): State beam -225 Phone exit beam -225 Insertion penalty 0
INFO: dict.c(320): Allocating 138824 * 20 bytes (2711 KiB) for word entries
INFO: dict.c(333): Reading main dictionary: /usr/local/lib/python2.7/dist-packages/pocketsphinx/model/cmudict-en-us.dict
INFO: dict.c(213): Dictionary size 134723, allocated 1016 KiB for strings, 1679 KiB for phones
INFO: dict.c(336): 134723 words read
INFO: dict.c(358): Reading filler dictionary: /usr/local/lib/python2.7/dist-packages/pocketsphinx/model/en-us/noisedict
INFO: dict.c(213): Dictionary size 134728, allocated 0 KiB for strings, 0 KiB for phones
INFO: dict.c(361): 5 words read
INFO: dict2pid.c(396): Building PID tables for dictionary
INFO: dict2pid.c(406): Allocating 42^3 * 2 bytes (144 KiB) for word-initial triphones
INFO: dict2pid.c(132): Allocated 21336 bytes (20 KiB) for word-final triphones
INFO: dict2pid.c(196): Allocated 21336 bytes (20 KiB) for single-phone word triphones
INFO: kws_search.c(406): KWS(beam: -1080, plp: -23, default threshold -225, delay 10)
2017-04-26 19:58:32 DEBUG: Setting up playback handler: SoxHandler
2017-04-26 19:58:32 INFO: Checking Internet Connection ...
2017-04-26 19:58:32 DEBUG: Starting new HTTPS connection (1): api.amazon.com
2017-04-26 19:58:33 DEBUG: https://api.amazon.com:443 "GET /auth/o2/token HTTP/1.1" 404 29
2017-04-26 19:58:33 INFO: Connection OK
2017-04-26 19:58:33 INFO: AVS token: Requesting a new one
2017-04-26 19:58:33 DEBUG: Starting new HTTPS connection (1): api.amazon.com
2017-04-26 19:58:34 DEBUG: https://api.amazon.com:443 "POST /auth/o2/token HTTP/1.1" 200 980
2017-04-26 19:58:34 INFO: AVS token: Obtained successfully
2017-04-26 19:58:34 DEBUG: Stopping audio play
2017-04-26 19:58:34 DEBUG: Stopped play. No SoX process to stop
2017-04-26 19:58:34 DEBUG: Playing audio: /opt/AlexaPi/src/resources/hello.mp3
Exception in thread Thread-1:
Traceback (most recent call last):
  File "/usr/lib/python2.7/threading.py", line 810, in __bootstrap_inner
    self.run()
  File "/usr/lib/python2.7/threading.py", line 763, in run
    self.__target(*self.__args, **self.__kwargs)
  File "/opt/AlexaPi/src/alexapi/triggers/pocketsphinxtrigger.py", line 70, in thread
    _, buf = inp.read()
ALSAAudioError: Capture data too large. Try decreasing period size

2017-04-26 19:58:34 DEBUG: Started play. sox -q /opt/AlexaPi/src/resources/hello.mp3 -t alsa default vol -6 dB pad 0 0
2017-04-26 19:58:36 DEBUG: Finished play.

Thank you very much for your patience @romain

Best regards,

Gerardo

romain · Apr 27, 2017, 7:49 AM

@borrigan Try to let “default” as an input device rather than myTest. The default thing will be in charge to call myTest itself (I think putting myTest directly doesn’t work for some weird reason I don’t now about)
Do you hear the “hello” when you run the script ?
If you still have the error after using the default, then ty to change you input handler from sox to vlc.I don’t think that’s related but at this point trying stuff might be the right thing x) Maybe that will fix it.

borrigan · Apr 27, 2017, 2:31 PM

@romain finally got it working by replacing myTest with default as you indicated. Than you so very much!! One short glitch though, sometimes Alexa’s phrases can not be heard completely. For example, at boot time I can hear “o” instead of “hello”. Same happens when talking to her.

But it is working!! Now time to test MMM-AlexaPi

Thanks again @romain

romain · Apr 27, 2017, 2:36 PM

@borrigan you’r welcom =)

borrigan · Apr 27, 2017, 6:35 PM

@romain quick question, I’ve noticed that Alexa takes a few seconds to answer, sometimes around 20 seconds. Does this happen to you as well? Could it be a slow internet connection. Thanks again!

romain · Apr 28, 2017, 10:28 AM

@borrigan It happen to me as well. MMM-alexa seem a little faster but it’s a pain to use so I rather wait few second for an answer