fork from rizal72/Cozmo-Voice-Commands
About Cozmo-What-Is-This
- Add a new command to Cozmo-Voice-Commands:
Cozmo, what is this
, then cozmo will tell you what it saw - Add new language: Chinese
- Cozmo, what is this
- I'm thinking! Oh, it is the coffee mug.
The project use Inception-v3 model to identify objects
Please note: cozmo what is this
only supports English/Chinese
Inception-v3 is trained for the ImageNet Large Visual Recognition Challenge using the data from 2012. This is a standard task in computer vision, where models try to classify entire images into 1000 classes, like "Zebra", "Dalmatian", and "Dishwasher"
About Cozmo-Voice-Commands
Issue multiple voice commands to Cozmo, and watch him execute all of them sequentially: highly customizable, you can add new commands with ease. Recognizes English, Italian, French, Dutch but it's very easy to add new languages!
Assuming that you've already performed the Cozmo SDK Setup, specific for your platform:
- requires
portaudio
:
-
on MacOS (see Homebrew if you don't know what
brew
is):
brew install portaudio
-
on Linux:
sudo apt-get install flac portaudio19-dev python-all-dev python3-all-dev && pip3 install --user PyAudio
-
on Windows:
you only need to install git as it is not included by default.
-
clone the project :
git clone https://github.com/wwj718/Cozmo-What-Is-This
-
install the requirements:
- cd Cozmo-What-Is-This
- pip install -r requirements.txt (python3.5 virtual env)
- python classify_image.py --model_dir tf/imagenset/
python cvc.py
- Optional arguments:
--version[-V]
: print version and exit
--no-wait[-N]
: enable deprecated continuous listening mode
--log[-L]
: enable verbose logging - choose speech recognition language and press enter.
- press SHIFT when you are ready, then issue your commands by voice (you have 5 seconds to start talking before it Timeouts), not too far from your PC, taking care to include the words "Cozmo" or "Robot" before any command you'll say: " COZMO, what is this? "