This is a hacked together prototype to enable creating a Google Doc with one or more Voice Memos from the Apple iPhone app. If there was an easier way to access Voice Memos, I'd be doing that.

My process is:

1. Record voice memos
2. Share them to a location on my laptop
3. Move them files to the /tmp project directory
4. Run this script with the file list and path to the directory like this: (.venv) /path/to/python /path/to/project/util.py -d /path/to/project/tmp -f file1.m4a file2.m4a file3.m4a

It relies on the OpenAI API and the Google API.

You'll need to create a .env file with the following entries:

CREDENTIALS_JSON=path/to/google/credentials/json
TOKEN_JSON=path/to/google/topen/json

OPENAI_ORGANIZATION=from your Open AI account
OPENAI_PROJECT=from your Open AI account
OPENAI_API_KEY=from your Open AI account

Follow these steps to get credentials.json from your GCP account

1. Go to the Google Cloud Console.
2. Create/Select your project.
3. Navigate to "APIs & Services" -> "Credentials."
4. Click "+ Create Credentials" -> "OAuth client ID."
5. Choose "Desktop app" as the application type.
6. Give it a name (e.g., "Voice to Doc").
7. Click "Create."
8. Download the credentials.json file and place it in your project /config directory.

token.json will be created and saved on first run of the script.

For OpenAI, follow the Developer quickstart here https://platform.openai.com/docs/quickstart.