TongueType

What is TongueType?

TongueType is a macOS voice dictation app that lets users hold a hotkey, speak, and release to insert transcribed text at the cursor. It runs Whisper AI locally on Apple Silicon Macs, so transcription happens on the device rather than in the cloud.

The app is designed for quick, hands-free text entry in places like email, chat, notes, prompts, and code comments. It also supports transcription from audio and video files, making it useful for converting recorded speech into text without uploading files to external services.

Key Features

• Local transcription with Whisper AI on Apple Silicon Macs, so audio stays on the device.
• Push-to-talk dictation workflow: hold a hotkey, speak, then release to insert text where the cursor is.
• Support for 12 languages, with automatic language detection.
• File transcription for WAV, MP3, MP4, and MOV inputs, which extends the app beyond live dictation.
• Menu bar app design with no Dock icon or main window, keeping it available without occupying workspace.
• Customization options for hotkey, activation grace period, double-tap latch mode, overlay appearance, and insertion behavior.
• Postprocessing rules for removing annotations like [music] or (laughter), canceling dictation with spoken phrases, and mapping spoken cues such as “new line” or “question mark.”
• iCloud sync for settings across Macs, so preferences can follow the user between devices.

How to Use TongueType

Install the app on a compatible Mac, then set or keep the default hotkey. When you need to dictate, hold the key, speak naturally, and release it to place the transcript into the current app at the cursor location.

For recordings, drag a supported audio or video file into TongueType and let it transcribe the contents locally. Users can also adjust settings for language behavior, appearance, and text cleanup before using it regularly.

Use Cases

• Writing emails or Slack messages faster than typing, especially when the user wants to speak instead of switching context to type.
• Drafting code comments, prompts, or short technical notes directly into an editor.
• Transcribing meeting recordings, interviews, or voice memos stored as WAV, MP3, MP4, or MOV files.
• Using dictation in a privacy-sensitive workflow where audio should remain on the Mac.
• Helping users who find keyboard input painful, slow, or impractical and need voice as an alternate input method.

FAQ

Is TongueType cloud-based?

No. The source says transcription runs locally on the Mac with Whisper AI, and audio does not leave the device.

Does TongueType support live dictation and file transcription?

Yes. It supports live push-to-talk dictation as well as transcription from supported audio and video files.

Which languages are supported?

The page says TongueType supports 12 languages and includes auto-detection, but it does not provide a complete language list beyond examples such as English, Spanish, French, German, Japanese, Chinese, and Korean.

Is there a free version?

Yes. The page states that the free version includes all features, with 30 minutes of live dictation per month and file transcription limited to the first 10 seconds of each file.

What platform does it run on?

TongueType is for macOS 14+ on Apple Silicon Macs.

Alternatives

• Built-in macOS dictation: a simpler system-level option for basic speech-to-text, but it does not appear to offer the same local Whisper-based workflow or the same customization described here.
• Cloud-based dictation apps: these may offer cross-platform access or account-based syncing, but they generally route audio through external services rather than keeping transcription fully local.
• General transcription tools: apps focused on converting recordings to text can be a fit if file transcription is the main need, but they may not be optimized for instant insertion into whatever app is currently active.
• Voice input extensions or accessibility utilities: these can help users who need voice as an alternative input method, though they may emphasize different workflows such as system-wide control, accessibility features, or note-taking rather than hotkey-driven dictation.