Configuring OpenAI Whisper for Transcription
Introduction
This document describes the steps you need to take to allow 3CX to transcribe your calls using OpenAI Whisper.
Setting up an OpenAI / Whisper Account
- Create a new project name
- Click on settings in the top right corner
- Set up “Billing” to complete the subscription process
- Click on “Usage Limits”
- Put a budget and Save
- Go to Project and create new project
- Select new project, click on “Settings” > “Limits”
- Configure “Monthly Budget”, “Notification threshold”, Add models: “whisper-1” and “gpt-4-turbo” and set “Rate limits”.
- Click on “Dashboard” > “API Keys” and press the “Start verification” to verify your phone number.
- After you receive SMS verification, you will be prompted to create an API key name, and set Permissions to “All”
- Copy and save your key somewhere safe as it will not be accessible again.
Setting up OpenAI / Whisper in 3CX
- Log in to the 3CX Admin Console and navigate to “Integrations” > “Open AI Whisper”.
- Check the option “Enable OpenAI Whisper”.
- Paste the api key into the secret key field (copied in Step 5 above).
- Select what you want to transcribe.
- Press “Save”.
- From now on your selected transcription will be made using OpenAI.
See also
- Voicemail & Recording Transcription using Google Cloud.
- Google Cloud Storage & Speech API V2 Transcription for Recordings/Voicemail
Last Updated
This document was last updated on 13 June 2024