WhisperDirect
Choose on-device Apple Speech (free, no API key) or the OpenAI Whisper API (high-accuracy, pay-as-you-go). Speaker diarization automatically identifies and color-codes multiple speakers. From recording to transcription, AI summaries, meeting minutes, and integrations with Slack, Google Drive, Discord, and Webhooks — the entire workflow is handled in one app. No subscriptions.
Download on the App Store
📌 Pay Only for What You Use
WhisperDirect does not charge any subscription. Use Apple Speech for free, on-device, or pay OpenAI directly for Whisper API — no middleman markup, no monthly fees.
Apple Speech (On-Device)
$0/forever
Transcription runs entirely on your device with iOS 26's Apple Speech. No API key, no network, no cost. Ideal for privacy-sensitive recordings.
Whisper API (OpenAI)
≈ $0.36/hour
Whisper API is priced at approximately $0.006/minute. With $5, you can transcribe about 14 hours of audio with high accuracy.
AI Summaries & Minutes
A few cents/run
Choose from GPT-4.1-nano, GPT-4.1-mini, GPT-5-nano, GPT-5-mini, or GPT-6 Luna. Even long texts of 1,000–2,000 words can typically be processed for just a few cents.
Available models may change based on OpenAI's offerings.
App Fee
One-time purchase
No monthly subscription. After a 7-day free trial, unlock once and use it forever. You only pay OpenAI directly for what you consume — nothing more.
📌 Get Started in 3 Steps
From setup to your first transcription in just a few minutes.
📌 Main Features
From audio capture to transcription, speaker diarization, AI summarization, and team sharing — everything you need to streamline your post-meeting workflow.
Two Transcription Engines
Switch between Apple Speech (on-device, free, no API key, iOS 26+) and the OpenAI Whisper API (high accuracy, pay-as-you-go) in Settings. Choose privacy and cost, or maximum accuracy — per your needs.
Speaker Diarization
Runs NVIDIA Nemotron 3 entirely on-device to automatically identify and color-code multiple speakers. Review conversations with per-speaker bubbles, rename speakers, and merge or reassign turns. ~200MB model, downloaded on first use.
Webhook Integration
Post transcripts, summaries, and meeting minutes to Discord, Slack, n8n, Make, Zapier, or any custom endpoint. Just set a URL. Header authentication (API key) is supported for custom webhooks.
Versatile Import Options
Record in real-time with the microphone button — including Bluetooth microphones, switchable via the selector above the record button. Import audio files via the Share Sheet, or load video files with automatic audio extraction and compression. Supported audio: mp3, m4a, aac, wav, flac, ogg, opus, wma, amr, mpga, webm, aiff, caf. Supported video: mp4, mov, m4v, webm, mkv, avi, mpeg, mpg.
Slack & Google Drive Integration
Automatically post transcripts, summaries, and meeting minutes to a specified Slack channel. Supports automatic backup to Google Drive. When a meeting ends, your notes are delivered to the team without any manual steps.
Redo & Change Language
Re-run transcription anytime with the current settings, or switch the recognition language from the detail view to reprocess automatically. Perfect for multilingual recordings.
Playback-Synced Highlighting
Transcript text highlights automatically in sync with audio playback. Jump to any segment with a tap. Timeline markers are configurable in 5-second steps, making it easy to navigate long recordings.
On-Device OCR (Free)
Extract text from photos of whiteboards or documents using high-accuracy OCR. Supports batch processing of multiple images. All processing happens on-device — no API cost.
Customizable AI Prompts
Edit the prompts used for summaries and meeting minutes directly from Settings. Tailor the AI output to your team's needs — extract action items only, output in bullet points, or any format you choose.
Flexible Export
Export audio files, transcripts, summaries, and meeting minutes. Also supports VTT and SRT subtitle file generation for use with video content and editing tools.
Real-Time Cost Estimation
Estimated API costs are calculated in real time based on audio length and character count, so you always have a clear sense of usage before you run a job.
Detailed Customization
Select your LLM model, adjust timeline intervals in 5-second steps, edit prompts, and configure Slack, Google Drive, and Webhook settings — all from a single settings screen.
📌 Built for These Workflows
Business Meetings
Record your meeting and generate AI meeting minutes immediately after. Speaker diarization keeps track of who said what. Auto-post to Slack to eliminate manual note-taking entirely.
Interviews & Research
Transcribe long interviews with high accuracy. Diarization separates interviewer and interviewee, and timeline markers make it easy to reference specific statements.
Lectures & Seminars
Record lectures and automatically extract key points for study notes and review materials.
Video Subtitles
Import a video file and export VTT or SRT subtitle files ready for YouTube or your video editing software.
🔑 Before You Get Started
- An OpenAI API key is required for Whisper API transcription and for AI summaries / meeting minutes. It is not required if you only use the on-device Apple Speech engine for transcription.
- Transcription and AI summarization costs are billed directly to your OpenAI account based on OpenAI's pricing.
- Speaker diarization (NVIDIA Nemotron 3) runs entirely on-device. About 200MB of model files are downloaded on first enable.
- Pricing, available models, and features may change based on OpenAI's offerings.
- Cost figures are based on OpenAI's published API pricing. Always check OpenAI's official pricing page for the latest rates.
- Free trial: 7 days included before purchase.
- Requires iOS 17.0 or later / iPadOS 17.0 or later / macOS 14.0 or later (Apple M1 or later). Apple Speech engine requires iOS 26.0 or later. Speaker diarization requires iOS 18.0 or later.
Supported Formats
Audio: mp3, m4a, aac, wav, flac, ogg, opus, wma, amr, mpga, webm, aiff, caf
Video: mp4, mov, m4v, webm, mkv, avi, mpeg, mpg