How accurate is Embedding Gemma Transcribe?

The developer says its word error rate is about 5%, higher than the 2-3% typical of leading speech recognition models, but expects further training to improve it.

What do I need to run it?

An Apple Silicon Mac running macOS 14 or later. The app is about 2.5MB, and the models it downloads are about 715MB.

Audio & SpeechOpen-Source AIGIGAZINE AI6m agoOct 15, 2026, 01:00 JST

Embedding Gemma Transcribe turns EmbeddingGemma 2 into a 3-second Mac transcriber

Send on LINE

3 Key Points

What happened
A free app called Embedding Gemma Transcribe runs offline on Apple Silicon Macs with macOS 14 or later. It transcribed a 15-minute-4-second Stanford speech in 3 seconds using about 715MB of models.
Why it matters
It shows an embedding model built to encode meaning, not to recognize speech, can be repurposed for transcription.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Get news like this every morning, free

FAQ

How accurate is Embedding Gemma Transcribe?: The developer says its word error rate is about 5%, higher than the 2-3% typical of leading speech recognition models, but expects further training to improve it.
What do I need to run it?: An Apple Silicon Mac running macOS 14 or later. The app is about 2.5MB, and the models it downloads are about 715MB.

GIGAZINE AIRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.