Guide
With your own API key, the app license and the provider's usage bill are separate. Estimate transcription from the amount of audio you send, then account for any AI cleanup.
Groq lists these audio transcription rates in its speech-to-text documentation. Check that page before budgeting because prices and limits can change.
| Model | Price per audio hour |
|---|---|
whisper-large-v3-turbo | $0.04 |
whisper-large-v3 | $0.111 |
Groq specifies a minimum billed duration of 10 seconds per request. Many very short recordings can therefore cost more than a calculation using recorded duration alone.
Thirty minutes of audio per working day over 22 days is 11 audio hours. At the rates above, transcription would be about $0.44 with Turbo or $1.22 with large-v3, before any minimum-duration effects or separate cleanup charges.
Formula: billed audio hours × your model's hourly rate. Check actual usage in your provider dashboard after a normal day's work.
VoiceForge is $79 once, including future updates. Optional cloud usage is paid to your provider. Local Whisper transcription needs no API key and has no provider usage bill; it uses your computer's resources.
Compare the total you expect to spend over the time you expect to use the app. A one-time license, a subscription, and a local tool can each make sense for different usage patterns.
See VoiceForge's local and cloud choices →
Ten practical ways to plan, dictate, and review a useful first draft.
No other email, ever.
Prices are Groq's published rates at the time of writing and can change. Check their docs for current numbers.
usevoiceforge.com