Nicla Voice Documentation

Purpose

This serves as a guide to help people who wish to setup Nicla Voice, for audio recording and processing purposes.

Setting up Nicla Voice

1. Download Arduino IDE and use the board manager/ library manager to install the default required libraries (Click Yes to install all when prompted).
2. Refer to https://docs.arduino.cc/tutorials/nicla-voice/user-manual to install additional firmware

Additional Libraries for Record and Stream

1. Refer to https://forum.arduino.cc/t/nicla-voice-record-audio/1095508 for basic background
2. Use the Arduino IDE ⇒ File⇒ NDP ⇒ Record_and_stream example within Arduino IDE.
3. Be sure to install the prerequisites libraries:
   • https://github.com/pschatzmann/arduino-libg722
   • https://github.com/pschatzmann/arduino-audio-tools/
4. Can place the libraries into ~\Documents/Arduino/libraries by following this guide https://roboticsbackend.com/install-arduino-library-from-github/
5. For arduino-audio-tools, be sure to change the version to v0.9.4 by using command git checkout v0.9.4 as latest version has breaking changes

Command to save recorded audio as .g722 file

After loading the program into Nicla Voice, you would need to pipe the captured audio from serial output to a file

1. Using terminal, list the used serial port via ls /dev/tty.*
2. Using updated screen program, run screen -L -Logfile arduino_output.g722 /dev/tty.<port-number> 115200

• example: screen -L -Logfile arduino_output.g722 /dev/tty.usbmodem00F1F5FF3 115200

3. You can stop the terminal by pressing ctrl + A and ctrl + \.

Open the .g722 file in Audacity

Using a program like Audacity, you can import the .g722 compressed file and play the audio