Transcribe audio to text with automatic speaker labels for recordings with multiple speakers, 100% on your own computer. No cloud, no uploads, no account. Your recordings never leave your machine. On-device AI starts instantly and uses a fraction of the energy a datacenter would burn.

Sign in to add this item to your wishlist, follow it, or mark it as ignored

This game is not yet available on Steam

Planned Release Date: Q4 2026

Interested?
Add to your wishlist and get notified when it becomes available.
 

About This Software

Your audio never leaves your computer

ChimpType turns speech into text and automatically labels who is speaking when a recording has multiple speakers, entirely on your own hardware. No cloud service, no file uploads, no account, no "we may use your data to improve our models." Drop in a recording and it stays on your machine from the first second to the final transcript.

That makes ChimpType a fit for the audio you can't just hand to a website: client calls, interviews, medical and legal recordings, journalists' source tapes, confidential meetings, and research data. It is private by architecture, not by policy. Unplug your network and ChimpType keeps right on working.

Green AI: a laptop, not a datacenter

Cloud transcription ships your audio to racks of power-hungry GPUs running around the clock. ChimpType does the same job on the computer already on your desk. A transcription model running on an Apple Silicon MacBook Air draws a tiny fraction of the electricity a datacenter node burns for the same work, with no idle server farm humming on standby waiting for your upload. Same transcript, a rounding error of the energy and carbon.

What it does

  • Accurate speech-to-text. Powered by OpenAI's Whisper, running locally on your GPU or CPU.
  • Multiple speakers, automatically labeled. Speaker diarization separates and labels each voice in a conversation, so you get a clean record of who said what.
  • Two modes. Quick whole-file transcription, or a full speaker-by-speaker breakdown with color-coded segments.
  • Edit everything. Fix a word, rename speakers, and tidy the transcript inline before you export.
  • Export anywhere. Plain text, SRT subtitles, or JSON.
  • Cross-platform. Windows, macOS (Apple Silicon), and Linux.
  • Fully offline. Once installed, ChimpType makes zero network calls.

Who it's for

Podcasters and video editors who need subtitles. Journalists and researchers transcribing interviews. Lawyers and clinicians handling confidential audio. Students turning lectures into notes. Anyone who wants a fast, accurate transcript without handing the recording to someone else's servers.

AI Generated Content Disclosure

The developers describe how their game uses AI Generated Content like this:

ChimpType is an AI application. Its core feature uses AI speech recognition and speaker diarization models, running entirely on your own device, to transcribe audio you provide and label who is speaking. Generative AI was also used to produce this store page's assets, including the artwork and the trailer. The trailer's narration was generated with ChimpChat, ChimpType's sister app.

System Requirements

Windows
macOS
SteamOS + Linux
    Minimum:
    • Requires a 64-bit processor and operating system
    • OS: Windows 10 64-bit
    • Processor: Quad-core Intel Core i5 / AMD Ryzen 5
    • Memory: 8 GB RAM
    • Graphics: NVIDIA GPU with 8 GB VRAM for GPU acceleration (CPU-only supported, slower)
    • Storage: 6 GB available space
    • Additional Notes: Runs 100% offline — your audio never leaves your machine. A CUDA GPU is recommended; CPU-only works but is slower.
    Recommended:
    • Requires a 64-bit processor and operating system
    • OS: Windows 11 64-bit
    • Processor: 8-core Intel Core i7 / AMD Ryzen 7
    • Memory: 16 GB RAM
    • Graphics: NVIDIA RTX 3060 or better (8 GB+ VRAM)
    • Storage: 10 GB available space
    Minimum:
    • Requires an Apple processor
    • OS: macOS 26
    • Processor: Apple Silicon (M1)
    • Memory: 8 GB RAM
    • Graphics: Apple Silicon GPU (MPS acceleration)
    • Storage: 6 GB available space
    • Additional Notes: Apple Silicon recommended for on-device GPU acceleration. Runs fully offline.
    Recommended:
    • Requires an Apple processor
    • OS: macOS 26
    • Processor: Apple Silicon (M2 or better)
    • Memory: 16 GB RAM
    • Graphics: Apple Silicon (M2 / M3)
    • Storage: 10 GB available space
    Minimum:
    • OS: Ubuntu 22.04 LTS 64-bit
    • Processor: Quad-core Intel Core i5 / AMD Ryzen 5
    • Memory: 8 GB RAM
    • Graphics: NVIDIA GPU with 8 GB VRAM (CUDA); CPU-only supported
    • Storage: 6 GB available space
    • Additional Notes: Self-contained build, runs fully offline. CUDA GPU recommended; CPU-only works but is slower.
    Recommended:
    • OS: Ubuntu 24.04 LTS 64-bit
    • Processor: 8-core CPU
    • Memory: 16 GB RAM
    • Graphics: NVIDIA GPU with 8 GB+ VRAM
    • Storage: 10 GB available space
Review Filters