On today's episode we dive into what it truly means to run AI locally on your Mac. When you upload your audio to a cloud service, you're essentially trusting strangers with your voice, your interviews, and your business conversations.
voxNest changes that. Everything — transcription, synthesis, even voice cloning — runs entirely on Apple Silicon through the power of MLX. No account to create, no audio leaves your device, and airplane mode works perfectly once models are downloaded.
We recorded this show on a MacBook Air M3, and Qwen3-ASR transcribed over half an hour of audio in just under three minutes. That's faster than real-time, with native punctuation and clean segment timestamps built right in.
00:00:01,500
On today's episode we dive into what it truly means to run AI locally on your Mac.
00:00:06,200
When you upload your audio to a cloud service, you're essentially trusting strangers...
00:00:12,800
voxNest changes that. Everything — transcription, synthesis, even voice cloning — runs entirely on Apple Silicon.
00:00:19,400
No account to create, no audio leaves your device, and airplane mode works perfectly.
00:00:27,100
Qwen3-ASR transcribed over half an hour of audio in just under three minutes.
00:00:31,500
That's faster than real-time, with native punctuation and clean segment timestamps.
00:00:01,500 → 00:00:06,000
On today's episode we dive into what it truly means to run AI locally on your Mac.
00:00:06,200 → 00:00:12,500
When you upload your audio to a cloud service, you're essentially trusting strangers with your voice.
00:00:12,800 → 00:00:19,200
voxNest changes that. Everything — transcription, synthesis, even voice cloning — runs on Apple Silicon.
00:00:19,400 → 00:00:27,000
No account to create, no audio leaves your device, and airplane mode works perfectly.
00:00:27,100 → 00:00:32,000
Qwen3-ASR transcribed over half an hour in under three minutes. That's faster than real-time.
