/amznas

Files and scripts for Amazonian Nasality project

Primary LanguagePythonOtherNOASSERTION

amznas

Files and scripts for Amazonian Nasality project

Installation

(64-bit Windows only)

  1. Open an Anaconda Prompt from the Start Menu.
  2. Clone the repository: git clone https://github.com/rsprouse/amznas
  3. Prepare the amznas Anaconda environment:
    • For first time creation: conda env create -f amznas\environment.yml
    • To update an existing environment: conda env update -f amznas\environment.yml --prune

Also, Recorder.exe from laryngograph.com must be copied to the C:\bin directory.

Running the amznas acquisition utility

  1. Open an Anaconda Prompt from the Start Menu.
  2. Activate the amznas environment: conda activate amznas
  3. Make an acquisition:
    • Unlimited duration: python amznas\amznas.py acq --researcher XXX --lang YYY --spkr ZZZ --item ITEM
    • Specified duration in seconds: python amznas\amznas.py acq --researcher XXX --lang YYY --spkr ZZZ --seconds 5 --item ITEM
    • (For Marina and Thiago, who have the older EGG-D800 device, add the parameter --dev_version 1)

Data

Recordings are stored in session folders under C:\Users\lingguest\Desktop\amznas. Session folders are created under the relative path ISO\SPK\YYYYMMDD, where:

  • ISO is the language ISO code
  • SPK is the speaker initials
  • YYYYMMDD is the date of the acquisition.

Within the session folders, filenames are of the form ISO_SPK_YOU_YYYYMMDDTHHHMMSS_ITEM_TOKEN, where:

  • ISO and SPK are the same as in the path
  • YOU is the researcher (you)
  • YYYYMMDDTHHMMSS is the date and timestamp of the acquisition
  • ITEM identifies the content of the acquisition
  • TOKEN is the instance of the ITEM in the session. Tokens are numbered automatically, starting at 0.

The Windows filesystem is not case-sensitive, so ITEM values that differ only in case are treated as identical when token numbers are calculated. Keep that in mind if your transcription system is case-sensitive.

Troubleshooting

If you get a ModuleNotFoundError, there is a good chance that you did not start an Anaconda Prompt or did not activate the amznas environment. Try repeating the steps in the 'Running the amznas acquisition utility' section.

If you get Error: No such option: <option>:

  1. Check to see if the option name was mistyped.
  2. Check to see whether you included a valid subcommand name (usually acq) to the script, e.g. python amznas\amznas.py acq ....