Whisper API costs 10x more than hosting an VM? (2024)

FAQs

How much does OpenAI Whisper API cost? ›

Whisper, the speech-to-text model we open-sourced in September 2022, has received immense praise from the developer community but can also be hard to run. We've now made the large-v2 model available through our API, which gives convenient on-demand access priced at $0.006 / minute.

Other third party providers currently price access to their Whisper models at up to $0.006 per minute — which seems cheap, but for a business like Mux, that $0.006 per minute isn't a trivial addition to our per-minute costs.

Whisper API, while not free forever, does offer generous free credits to new users.

The price of whisper is $0.006 per minute, or $0.0001 per second (rounded to seconds per pricelist).

By default, the Whisper API only supports files that are less than 25 MB. If you have an audio file that is longer than that, you will need to break it up into chunks of 25 MB's or less or used a compressed audio format.

There is no “free account” for API. The use of the service costs money by the amount of data used. There is only possibility of a free trial credit, which expires three months after you first created your OpenAI account. Now you'll need to purchase a credit balance in order to make calls.

With Deepgram, you'd save roughly 40% with higher overall quality and faster turnaround for comparable audio. Unlike Whisper, we can also package the API for use in VPC or on-prem applications.

OpenAI's Whisper-v2, the most accurate Whispers, has a median WER of 8.06% and takes 10-30 minutes on average to transcribe one hour of audio. As notes previously, a big advantage with Whisper is that the model comes in various sizes, enabling developers to strike the right balance between speed and accuracy.

Whisper is a machine learning model for speech recognition and transcription, created by OpenAI and first released as open-source software in September 2022.

Whisper Model via Azure AI Speech might be best for: Transcribing files larger than 25MB (up to 1GB). The file size limit for the Azure OpenAI Whisper model is 25 MB.

Do I need an API key for Whisper? ›

OpenAI Account: You need an active OpenAI account to access the Whisper API. API Key: Obtain your OpenAI API key from the platform. This key will authenticate your requests to the Whisper API.

Whisper has two parts to its advertising business: direct sales for advertising clients, and programmatic ads that show up on a user's phone when they use the Whisper app. Whisper has experimented with advertising by working with Fox Studios, MTV, Universal, Hulu and FX on customized branded campaigns.

Referencing the Open AI pricing page Each minute costs $. 006. This works out to $. 36 cents an hour for one user.

Google Cloud Speech-to-Text

Google's deep learning neural network algorithms are the most advanced in automatic speech recognition (ASR).

Conclusion. By following these steps, you can run OpenAI's Whisper locally and transcribe audio files without needing to call the OpenAI API. This setup allows you to have full control over the transcription process and avoid any potential latency or costs associated with using the API.

See our Pricing page for details. Is Whisper still free in the playground? Starting March 1st, 2023, with the Whisper API launch it is no longer free in the playground.

Price: $20 per month. Availability: Web or mobile app. Features: Voice recognition; memory retention; multiple GPTs to choose from. Image generation: Yes.