MacWhisper 12 brings first-ever on-device speaker recognition to Mac transcription

MacWhisper 12 introduces a groundbreaking advancement in AI transcription technology with its new automatic speaker recognition feature that works entirely on-device. This development represents a significant milestone for content creators, journalists, and professionals who rely on accurate transcription, as it addresses one of the most challenging aspects of audio-to-text conversion while maintaining privacy and security through local processing.

The big picture: MacWhisper 12 has become the first Mac application to offer automatic speaker recognition that works entirely on-device, solving a major challenge in transcription technology.

• The app can now automatically detect different speakers in audio files, group their speech, and label them accordingly without sending any data to external servers.
• This privacy-focused approach processes all transcription locally on the user’s Mac, making it accessible even without an internet connection.

Key details: The new feature emerged from a collaboration between developer Jordi Bruin and ArgMax, utilizing their WhisperKit Pro and SpeakerKit models.

• Users can simply transcribe their audio and then assign names to each automatically detected speaker to create clearly organized transcripts.
• The update also works with ElevenLabs and Deepgram transcription APIs for additional speaker recognition support options.

What they’re saying: “This is a game-changer, and we’re proud to be the first to bring local, on-device speaker recognition to the Mac,” noted the developers regarding the breakthrough capability.

Accessibility and pricing: MacWhisper is available as a free download with basic OpenAI Whisper transcription included.

• Users can unlock the full capabilities, including the new on-device automatic speaker recognition, by purchasing MacWhisper Pro.
• A limited-time 15% discount is available with the code “FINALLY” to celebrate the release of this highly-requested feature.

Why this matters: Automatic speaker recognition significantly streamlines the workflow for anyone working with interview transcriptions, meeting notes, or multi-person conversations, eliminating the time-consuming process of manually identifying and labeling different speakers.