Last updated 07-28-2026
Category:
Reviews:
Join thousands of AI enthusiasts in the World of AI!
WhisperAPI
WhisperAPI turns uploaded audio and video files into text through a hosted transcription API built on OpenAI's Whisper model. You can send files via REST API calls or drag them into a browser dashboard, then download transcripts as JSON, plain text, VTT, DOCX, or PDF. Signup includes 5 free credits with no credit card required.
Most transcription APIs either lock you into monthly seats or expect you to supply your own OpenAI key. WhisperAPI runs self-hosted Whisper models on infrastructure it controls, sells pay-as-you-go credits that never expire, and shows the exact credit cost before each job based on model size and file length. That makes it a better fit for sporadic batch work than flat subscription tools where you pay the same fee whether you transcribe ten minutes or ten hours.
Developers can hit api.whisper-api.com with direct uploads or remote URLs, pick Whisper model sizes for speed versus accuracy, and wire transcriptions into Zapier workflows. Podcasters, video editors, and teams that need captions without maintaining GPU infrastructure can do the same work from the dashboard without writing code.
5 free transcription credits at signup with no credit card required
Self-hosted Whisper models including large-v2 with no OpenAI API key needed
Upload audio and video files up to 10GB with a Pro subscription
98+ languages with automatic spoken-language detection
Export transcripts as JSON, TEXT, VTT, DOCX, or PDF
Uploaded files auto-delete after 24 hours; audio is never used for model training
Zapier triggers fire when a transcription completes for workflow automation
No OpenAI API key needed because WhisperAPI hosts its own Whisper models.
Pay-as-you-go credits never expire, with bundles starting at $5 for 20 credits.
Both a REST API and a browser dashboard cover developers and non-coders.
Uploads up to 10GB with a Pro subscription and no per-minute transcription caps.
Uploaded files auto-delete after 24 hours and are never used for model training.
10GB file uploads require a Pro subscription rather than the free tier.
Credit cost varies by Whisper model size, speaker diarization, and file length.
No dedicated mobile app; access is through the browser or API only.
Does WhisperAPI require an OpenAI API key?
No. WhisperAPI hosts its own self-hosted copy of the open-source Whisper model and generates an API key for you at signup. Your audio is processed on WhisperAPI infrastructure and is not sent to OpenAI for training.
How much does WhisperAPI cost?
WhisperAPI uses pay-as-you-go credits with no monthly fee. You get 5 free credits at signup, then bundles start at $5 for 20 credits ($0.25 each). The 100-credit pack is $20, and larger volumes drop to $0.10 per credit on 1,000+ credit orders.
Do WhisperAPI credits expire?
No. WhisperAPI credits never expire after purchase. You can buy credits anytime and spend them whenever you need a transcription, whether that is days or months later.
What file formats does WhisperAPI accept?
WhisperAPI supports common audio and video formats including MP3, WAV, MP4, and M4A. With a Pro subscription you can upload files up to 10GB in size, and there are no per-minute caps on transcription length.
How long does WhisperAPI keep uploaded files?
WhisperAPI automatically deletes uploaded audio and video files after 24 hours. Only the transcription text stays in your account, which limits how long raw media sits on WhisperAPI servers.
What languages does WhisperAPI support?
WhisperAPI supports 98+ languages, including English, Spanish, French, German, Chinese, and Japanese. The service detects the spoken language automatically and does not require you to set a language code unless you want to.
Can I use WhisperAPI without coding?
Yes. WhisperAPI offers a browser dashboard with drag-and-drop uploads, live progress tracking, and one-click downloads in multiple formats. Developers can use the REST API at api.whisper-api.com, but the dashboard covers the same transcription workflow.