MacWhisper Now Supports Automatic Speaker Recognition
Upgrading MacWhisper with NVIDIA Parakeet & Automatic Speaker Recognition
Inspired by this Hacker News conversation, I decided to upgrade MacWhisper to try out NVIDIA Parakeet along with the new Automatic Speaker Recognition feature.
---
Test Results
I ran the upgraded setup against this 39.7 MB `.m4a` audio file from my Gemini 3 Pro write-up.
Performance:
It worked very well, producing clean transcripts and correctly identifying speakers.

---
Exporting Transcripts with Speaker Data
You can export the transcript with:
- Timestamps
- Identified speaker names
Steps:
- Go to `Share`
- Select `Segments`
- Choose `.json` as the export format

📄 Example output:
Here’s the resulting JSON file.
---
Monetizing AI-Generated Transcripts
If you’re looking to turn AI-generated transcripts into monetizable, cross-platform content, consider using AiToEarn.
Features of AiToEarn:
- Publish & manage AI content across multiple platforms:
- Douyin
- Kwai
- Bilibili
- Rednote (Xiaohongshu)
- YouTube
- X (Twitter)
- …and more
- Open-source ecosystem (GitHub repo)
- Built-in tools for:
- AI content generation
- Publishing automation
- Analytics & performance tracking
- Model ranking for best AI output
Goal: Efficiently monetize AI-driven creativity.
---
✅ This workflow shows how combining transcription tools with open-source publishing ecosystems can streamline both content production and monetization.