EventGhost

Sem;colon wrote:Hi Pako, IntelliSpeech doesn't work with EG 0.5.0 yet

I'm sorry, but now I do not have time to do it.
I did not install 5.0 so far.

Pako

The new version (0.6) of the plugin is available.
No innovations, only forced modifications (to regain functionality).

Pako

When trying to test the microphone speech to text I get an error as defined below

Traceback (most recent call last) (1710):
File "C:\Program Files (x86)\EventGhost\plugins\IntelliSpeech\__init__.py", line 2034, in run
r = sr.Recognizer(language = self.lng)
TypeError: __init__() got an unexpected keyword argument 'language'

This is a great plugin and would love to get this to work. Your help is greatly appreciated.

dorr35 wrote:r = sr.Recognizer(language = self.lng)
TypeError: __init__() got an unexpected keyword argument 'language'

It appears that you have the wrong version of some file. Please carefully follow all the steps to upgrade again.
Maybe it's my fault. I found that the ZIP file IntelliSpeech_0.6.zip was not assembled correctly.
There lacked the main directory IntelliSpeech. I added it in there and I did reupload.

Thank you for your reply. I replaced the folder with the new download but that did not fix the problem. The error has changed. When I activate mic to text the error is now

Exception in thread Thread-308:
Traceback (most recent call last):
File "threading.pyc", line 532, in __bootstrap_inner
File "threading.pyc", line 484, in run
File "C:\Program Files (x86)\EventGhost\plugins\IntelliSpeech\__init__.py", line 1841, in Animate
while self.run:
File "wx\_core.pyc", line 14619, in __getattr__
PyDeadObjectError: The C++ part of the Animation object has been deleted, attribute access no longer allowed.

Maybe I am implementing it wrong. I am fairly new to eventghost.

<?xml version="1.0" encoding="UTF-8" ?>
<EventGhost Version="1710">
<Macro Name="dictate" Expanded="True">
<Event Name="Intellispeech.dicate" />
<Action>
IntelliSpeech.MicToText([u'English', u'United States'], 2500, 0.80000000000000004, u'Success', u'Failure', True, u'Default input device', 1)
</Action>
</Macro>
</EventGhost>

Could you possibly show an example of the correct configuration?

I am running Windows 10 x64 with EventGhost 1.4.1.r1710

Thank you in advance for your reply.

Well, at least some progress. The error is changed, therefore, incorrect installation was confirmed.
So now we deal with a new bug.
I do not know how to get the same behavior for me - for me it works.
However, I tried to make some adjustments. Attached is a test version of the plugin.
Please try to replace __init__.py file in the root directory of the plugin.

Pako

Again thank you for your assistance. The above attached file was not successful. I am still receiving errors. I am pretty sure this is a problem with configuration. Would it be possible you could upload a working tree for me to look at? I have used speakeasy but this would allow me to mic to text which I could surely use.
Thank you again in advance for your assistance in this matter.

dorr35 wrote:... I am still receiving errors. I am pretty sure this is a problem with configuration.

I am afraid that it is not so.
I think it will rather cover the operating system and the like.
However, we can try it differently. Give your configuration file as an attachment and I will test how it behaves in my computer.

Pako

Here is an example of a configuration file I am currently working on. Once I quit speaking it gives this error in eventghost:

Exception in thread Thread-6:
Traceback (most recent call last):
File "threading.pyc", line 532, in __bootstrap_inner
File "threading.pyc", line 484, in run
File "C:\Program Files (x86)\EventGhost\plugins\IntelliSpeech\__init__.py", line 1841, in Animate
while self.run:
File "wx\_core.pyc", line 14619, in __getattr__
PyDeadObjectError: The C++ part of the Animation object has been deleted, attribute access no longer allowed.

I am hoping to get this working so I will be able to speak my searches in the text boxes inside Kodi search. Looking forward to your having a look at it and giving me your input.

It's exactly as I expected.
In my computer, it works normally. I have a Windows 8.1 / 64 bits.
However, I think that even so you can use it.
Only you have to turn off the animated indicator:Pako

Thank you. That removed the error but when I talk there is no text output. I tried to speak while notepad was open but nothing happened. The page remained blank. I assume I am missing a step somewhere? I know we're close to having this operational. I am sorry to keep bothering you but I really want this to work.

dorr35 wrote:I tried to speak while notepad was open but nothing happened.

I do not know what you expect. You do not need a notepad.
Try using the attached configuration file (it works as a translator from English into German).
Highlight the item "Microphone input" and press green button "Execute Item". Then speak.

Note: The result of processing input from the microphone is always an EVENT. In the picture is marked.
The text is a payload. It is necessary to somehow process.Pako

When I attempt to load the plugin I am getting the following errors:

Error loading plugin file: C:\EventGhost\plugins\IntelliSpeech
Traceback (most recent call last) (0.5.0-beta5):
File "C:\EventGhost\plugins\IntelliSpeech\__init__.py", line 114, in <module>
from _pyaudio import pyaudio
File "C:\EventGhost\plugins\IntelliSpeech\lib\_pyaudio\__init__.py", line 1, in <module>
from pyaudio import __version__
File "C:\EventGhost\plugins\IntelliSpeech\lib\_pyaudio\pyaudio.py", line 116, in <module>
import _portaudio as pa
ImportError: DLL load failed: The specified module could not be found.

Maybe it's because you're using EventGhost 5.0.
Please try EventGhost 0.4.1.r1722.
Unfortunately, I do not know how to make support for 5.0.

Pako

Hello,

I just saw this thread, unfortunately it's a bit late as I spent a part of the past week-end setting up the speech function using nirCmd and windows TTS.
Everything is working as long as I have a remote session on my eg server. Some minutes after having closed the rdp, the speech macros won't play any sound. No error in the log. This server is also used as a music player using other apps (spotify web client, squeeze player, etc.), all these apps are working flawlessly when TTS events keep silent.

Is there any clue to explain this? Should I migrate to this IntelliSpeech plugin to solve this?