Speech to text
Whisper Large v3
Multilingual speech-to-text — 99 languages, including Nepali.
- Status
- Live
- Languages
- 99 languages
- License
- MIT (OpenAI Whisper open release)
Model ID
| Model ID | Designed for |
|---|---|
whisper-large-v3 | Multilingual transcription and English translation |
Confirm access with GET /v1/models, then follow the utility-model docs for the canonical v1 REST or native gRPC request contract.
Pricing
whisper-large-v3: Free for a limited time. $0.00005 $0.00 per audio second
Overview
Whisper Large v3 is the third-generation release of OpenAI’s open Whisper model — robust, multilingual speech recognition trained on 680k+ hours of multilingual and multitask supervised data from the web.
Whisper Large v3 is available through two canonical v1 endpoint shapes and matching native gRPC methods:
/v1/audio/transcriptions— OpenAI-compatible multipart upload with language detection and optional timestamps where supported./v1/audio/translations— translate supported source languages directly into English.
When to pick it
- Anything an agent needs to listen to: voice messages, calls, voice notes, meeting recordings, podcast clips
- Multilingual workflows where you cannot pin the input language ahead of time
- Pipelines that chain transcription into a language-model call under the same Tarka account
Use model discovery and your organization limits as the source of truth for
access, upload size, and concurrency. Pass language=ne when the
source audio is Nepali.