Whisper Web

What it does

Whisper Web is a browser tool for turning speech into text with OpenAI’s Whisper models. Start in a desktop or mobile browser, then provide audio by microphone, file upload, or URL/browser audio. Free local mode processes audio in the browser and supports multiple languages with automatic language detection. Transcripts can be exported as TXT, SRT, VTT, or JSON. The stated audience is broad; its site describes the tool as being for everyone, and its homepage mentions developers and creators.

Free mode supports files up to 20 minutes and 200 MB, has no daily file-count limit, and requires no account. After model files are downloaded on first use, the site says this mode can work offline. Whisper Web Unlimited is an optional cloud workflow for longer files and batch uploads: up to 10 hours or 5 GB per file, with up to 50 files at once. Selected audio is uploaded for cloud transcription; Unlimited also lists cross-device sync and cloud storage. Use a browser supporting WebGPU for the advertised accelerated local inference; the site does not document specific browser versions or device requirements.

What it can do

• Transcribes microphone audio
• Transcribes uploaded audio files
• Accepts URL or browser audio
• Local speech recognition using Whisper models and WebGPU
• Automatic language detection across supported languages
• Exports transcripts as TXT, SRT, VTT, or JSON

Works with

• OpenAI Whisper models
• Transformers.js

Before you choose

• Free local files are limited to 20 minutes and 200 MB each. Unlimited supports up to 10 hours or 5 GB per file.
• Free local mode keeps audio in the browser, but model files are downloaded over the network on first use; Unlimited uploads selected files for cloud processing.
• Offline use is documented for the Free plan after model download; cloud processing requires an internet connection. The site does not specify supported browser versions or hardware.

Listed under

• Productivity
• Audio-to-text
• Lectures
• Meetings
• Podcasts
• Privacy
• Speech-to-text

Get started

Free local mode requires no account. Unlimited is an optional paid plan.

1. Open whisperweb.dev in a desktop or mobile browser that supports WebGPU for accelerated local processing.
2. For free local use, provide microphone audio, upload a file, or use URL/browser audio; no account is required.
3. For offline local use, allow model files to download on first use; choose Unlimited if you need its cloud workflow for longer files or batch uploads.

Local transcription is free without an account. Optional cloud Unlimited costs USD20/month or USD120/year, equivalent to USD10/month billed yearly. Monthly and annual subscriptions are alternatives; cloud uploads, storage and sync have different data handling from local use.

Source: whisperweb.dev/pricing, checked 2 Nov 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 2 Nov 2026. Each source supports part of the overview; availability and prices can change.

• Whisper Web - Free AI Speech Recognition | Browser-Based Transcription ↗whisperweb.dev
• About Us - Whisper Web | Privacy-First AI Speech Recognition | Whisper Web ↗whisperweb.dev
• Simple, Transparent Pricing | Whisper Web ↗whisperweb.dev
• Whisper Web - Free AI Speech Recognition | Browser-Based Transcription ↗whisperweb.dev

Company website

whisperweb.dev: product, pricing and policy pages

YC directory mirror

Batch and company name

Human review

Checked by a person

knowledge_base

manual_review

Last checked 2 Nov 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.