ImportError: No module named 'bin'  

  RSS

johnsonld123
Active Member
Joined: 6 months  ago
Posts: 6
31/10/2016 4:02 am  

Hello all

I am trying to get this setup on lubuntu 16. I think I have all my dependencies but receive this error when I first run ./assist.sh

 

Traceback (most recent call last):
  File "/usr/share/openassistant/assistant.py", line 5, in <module>
    from bin import core
ImportError: No module named 'bin'

Any suggestions on where I went wrong


ReplyQuote
OpenAssistant
Eminent Member
Joined: 7 months  ago
Posts: 27
31/10/2016 10:53 am  

Greetings John!

Thanks for trying out this early version of Open Assistant.

Please download and try again:

https://github.com/vavrek/openassistant

We've been working on changing the directory structure and this has created your errors.

Please let us know how things go!


ReplyQuote
johnsonld123
Active Member
Joined: 6 months  ago
Posts: 6
03/11/2016 12:17 am  

Thank you for responding and an awesome AI. I cannot wait to see all the future developments and hope to contribute.

I have tried the new download and can set Stella to start listening, but after starting I do not receive a response just

-warp_type        inverse_linear    inverse_linear
-wbeam            7e-29        7.000000e-29
-wip            0.65        6.500000e-01
-wlen            0.025625    2.562500e-02

< ? > up
< ? > up
< ? > if i
< ? > up up
< ? > the
< ? > up up
< ? > fuck
< ? > look up the op
< ? > for the
< ? > for
< ? > fuck up up up
< ? > up
< ? > if the op
< ? > if for up
< ? > pipe
< ? > the the app
< ? > up up
< ? > the
< ? > yep

I think I am missing something.

Thank you again


ReplyQuote
OpenAssistant
Eminent Member
Joined: 7 months  ago
Posts: 27
03/11/2016 8:39 pm  

lol ! What a nasty machine!

Seems like some kind of problem with your microphone at this point.

Stella currently tends to hear words like 'up' and 'fuck' when there are noises happening like taps or clicks.

As a test, I would make sure you can record audio using a program such as Audacity and that your sound is good.

I should really take the word 'fuck' out of the commands file.... Could really be shocking for some people. 🙂


ReplyQuote
johnsonld123
Active Member
Joined: 6 months  ago
Posts: 6
03/11/2016 9:51 pm  

Whoa whoa whoa...no need to remove it ( I like my AI fiesty). Tried recording and playback which was successful with Audacity. I am distro hopping so I also tried on Linux Mint 18 with the same results. I think I'm missing a step or something. Gonna run through the dependences.


ReplyQuote
OpenAssistant
Eminent Member
Joined: 7 months  ago
Posts: 27
05/11/2016 8:03 pm  

Hey John!

Try this ... At a command prompt type 'cat /proc/asound/cards'

Also try 'arecord -l'

These should give you a card number for your microphone / sound card.

Make sure that in 'assist.sh' the last command has the appropriate card number set like this:

python3.5 assistant.py -c -H 20 -m <cardnumber>

The one that works for me is:

python3.5 assistant.py -c -H 20 -m 0


ReplyQuote
birdperson26
New Member
Joined: 3 months  ago
Posts: 1
27/01/2017 4:29 am  

im using ubuntu . And i have the same problems but NOT with my selfmade speechrecognition .Im going to install Arch and try it on this distro . If it doesnt work there too then i call this a stupid hoax and just do it by myself.Also the wiki http://openassistant.org/wiki/doku.php looks rly fishy there wants somone rly desperate traffic on his website. So just test it on Arch.


ReplyQuote
  
Working

Please Login or Register