# Strange behaviour of wakewordtest

**URL:** https://community.openconversational.ai/t/strange-behaviour-of-wakewordtest/11907
**Category:** Archive - Mycroft Project
**Created:** [February 16, 2022, 11:13am UTC](https://community.openconversational.ai/t/strange-behaviour-of-wakewordtest/11907 "2022-02-16T11:13:58Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![TheRealDGD](https://community.openconversational.ai/user_avatar/community.openconversational.ai/therealdgd/32/3674_2.png) [@TheRealDGD](https://community.openconversational.ai/u/TheRealDGD)
#### Post date: [February 16, 2022, 11:13am UTC](https://community.openconversational.ai/t/strange-behaviour-of-wakewordtest/11907/1 "2022-02-16T11:13:58Z")

</div>

While trying to generate an new wake-word using precise, I discovered a strange behaviour of **mycroft-start wakewordtest**

I already fixed a missing function in FileMockMicrophone so now it’s running.  
But with the original Code from dev branch none of my wake words are recognized.  
When I insert some sleep statements at different places in wake\_word\_test.py more wakewords are recognized.  
It’s stange that a copy of a wave file is only recognized on the second try.

```auto
Waiting 2s for backend to start up
..
Wake word Not found - hey-jarvis-00.wav
2022-02-16 10:57:53.354 | INFO | 25169 | mycroft.session:get:72 | New Session Start: 64cdb4a7-5a1c-4a6b-9929-d60d0a68f2da
Wake word Detected - hey-jarvis-00a.wav

```

I think, there’s a timing issue because of asynchronus processing.  
It’s the same on my Raspi and on my Deskop.

I am happy to receive any suggestions.