README.md

Instruction

This code are using python3 and ubuntu.

1. First you need pull code to your environment or something else method, but you must create virtualenv. I used PyCharm(Preferred).

• git pull https://github.com/Firemanpl/Teams-speech-recognition-audio-stream.git

2. Install python3-dev:

• sudo apt-get install python3-dev

3. Install requirements.txt :

• pip install requirements.txt

4. Change view grid to list in settings on Teams:

5. Install pulseaudio. I used instruction from this link: https://www.youtube.com/watch?v=H4M1tn7LV8U
6. Create a virtual card in pulseaudio:

• put below command to terminal: sudo gedit /etc/pulse/default.pa
• copy and paste these lines to deafult.pa:

load-module module-null-sink sink_name=Output_to_Input sink_properties=device.description=Output_to_Input
load-module module-null-sink sink_name=Input_to_Output sink_properties=device.description=Input_to_Output
load-module module-loopback sink=Input_to_Output
load-module module-loopback sink=Output_to_Input

7. Enter your data to config.jsonSet time to a one -two minute before example meeting time.You must remember about correct format time like that:09:01 instead 9:1 .Otherwise, you will get error in python console. Record your voice/voices and paste to your project folder. Then copy name your filename your_example_voice.wav. You can paste infinity voices to config. Files must be formatted .wav !!! At the end set your language. For example, if your language/dialect is British English, it is better to use "en-GB" as the language rather than "en-US":

  {
    "email": "[email protected]",
    "password": "password",
    "current_members_less_than": 2,
    "teamname": "Test",
    "meetingtime": "22:10",
    "your_voice": ["record (2).wav" ,"record.wav","alert.wav"],
    "trigger_words": ["Json", "Manfreddy", "Stinky", "Rock", "example", "example", "example"],
    "language": "pl-PL"
  }

8. Start main.py: python3 main.py or green play sign in PyCharm. After run and join to online meeting say one with trigger_words.
9. Go to pulseaudio and check setting like this: (You have repeat these action every time.)

(Optional, not tested) Pyaudio on Windows.

• pip install pipwin

• pipwin install pyaudio