Subvocal:用雷达眼镜读取默读口型,输入速度超 200 wpm
People don't hate dictating. They hate dictating with other people around. I've been talking to my computer for a year now, it's way faster than typing. Then when I sit down on a train or in a café and go straight back to the keyboard, because nobody wants to hear my Slack replies. Subvocal fixes exactly that. You mouth the words, radar in the glasses reads your speech muscles, and text shows up at 200+ wpm wherever you type. But it only picks up what you deliberately mouth, so it isn't always listening. Third way to talk to a computer. Big if it works this well in the wild. Excited to get my pair in Q1 2027
Subvocal 通过眼镜中的雷达读取用户默读时的口型与语音肌肉动作,在任意输入框中以 200+ wpm 输出文字。它只识别用户刻意默读的内容,不会持续监听。作者期待在 2027 年 Q1 拿到自己的设备,并认为这是与电脑对话的第三种方式。
粉丝 25万
Everything AI + Robotics, deep dives, hottakes. I help business and individuals overcome in the age of AI. Optimist [email protected] for partnerships 🤝
- 浏览
- 4010
- 点赞
- 41
- 评论
- 18
- 分享
- 4
- 收藏
- 20
数字来自平台,为最近一次抓取时的数据。
来源:X · Linus Ekenstam · x.com