OpenAI Whisper Models by OpenAI
4 models2022–2024
Details
ResearcherOpenAI
LicenseProprietary
Commercial useCommercial use: conditional
Models4
Released2022–2024
Capabilities
MultimodalAll models
Links
WebsiteAbout
OpenAI's Whisper family provides multilingual automatic speech recognition and translation models exposed through the OpenAI Audio API.
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
4 in view
Whisper large-v3-turboCurrent
Use when the workload needs speech recognition, multimodal inputs, and audio.
2024-01speech recognitionmultimodal inputsaudio
Whisper BaseCurrent
Use when the workload needs speech recognition, 74M parameters, and multimodal inputs.
2022-12speech recognition74M parametersmultimodal inputs
Whisper Tiny ENCurrent
Use when the workload needs speech recognition, 39M parameters, and multimodal inputs.
2022-12speech recognition39M parametersmultimodal inputs
WhisperCurrent
Use when the workload needs speech recognition, multimodal inputs, and audio.
2022-09speech recognitionmultimodal inputsaudio
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Whisper large-v3-turbo | Use when the workload needs speech recognition, multimodal inputs, and audio. | 2024-01 | speech recognitionmultimodal inputsaudio | Current |
| Whisper Base | Use when the workload needs speech recognition, 74M parameters, and multimodal inputs. | 2022-12 | speech recognition74M parametersmultimodal inputs | Current |
| Whisper Tiny EN | Use when the workload needs speech recognition, 39M parameters, and multimodal inputs. | 2022-12 | speech recognition39M parametersmultimodal inputs | Current |
| Whisper | Use when the workload needs speech recognition, multimodal inputs, and audio. | 2022-09 | speech recognitionmultimodal inputsaudio | Current |
Release Timeline
3 release groups2024-01
1 current
Whisper large-v3-turbo
Currentspeech recognitionmultimodal inputsaudio
2022-12
2 current
Whisper Base
Currentspeech recognition74M parametersmultimodal inputs
Whisper Tiny EN
Currentspeech recognition39M parametersmultimodal inputs
2022-09
1 current
Whisper
Currentspeech recognitionmultimodal inputsaudio
Specifications(4 models)
| Model | Released | Parameters | Multimodal |
|---|---|---|---|
| Whisper large-v3-turbo | 2024-01 | — | Yes |
| Whisper Base | 2022-12 | 74M | Yes |
| Whisper Tiny EN | 2022-12 | 39M | Yes |
| Whisper | 2022-09 | — | Yes |
Frequently Asked Questions
- What is OpenAI Whisper used for?
- OpenAI Whisper is used for audio, speech recognition, and vision and multimodal work. The family description and listed model capabilities point to those workloads as the best fit.
- How does OpenAI Whisper compare to GPT Realtime 2?
- OpenAI Whisper by OpenAI is strongest where you need audio, while GPT Realtime 2 by OpenAI is the closest related family to check for realtime voice. OpenAI Whisper has 4 listed variants, while GPT Realtime 2 reaches up to 131k context, so compare the specs and pricing tables before choosing a production model.
- Which OpenAI Whisper model should I use?
- If price is the main constraint, use the pricing table first because OpenAI Whisper does not have complete provider pricing in the local data. For the most capable/latest local choice, evaluate Whisper large-v3-turbo with multimodal inputs.
