Yap

Product Introduction

1. Definition: Yap is a native macOS desktop application for real-time, on-device speech-to-text dictation. It is a lightweight utility built in Swift that leverages Apple's Speech framework.
2. Core Value Proposition: Yap exists to provide a fast, private, and offline-first dictation tool that integrates seamlessly into any text field on macOS, eliminating the need for cloud-based transcription services, API keys, or subscriptions.

Main Features

1. On-Device Speech Recognition: Yap processes audio entirely locally using macOS's built-in Speech APIs (Speech framework). This means no large AI models need to be downloaded, and all audio processing happens on the user's Mac's Neural Engine or CPU, ensuring zero data transmission to external servers.
2. System-Wide Hotkey Activation: The application operates via a single, user-configurable global keyboard shortcut. Pressing the hotkey begins recording speech from the default microphone; pressing it again stops recording and instantly pastes the transcribed text into the currently focused text field, whether it's a browser, code editor, terminal, or email client.
3. Local Transcript History: Every dictation session is automatically saved to a local database on the Mac. This history is searchable, allowing users to quickly find and reuse previous transcripts without any cloud sync, enhancing productivity and maintaining privacy.
4. Offline-First Architecture: The app contains no network code. All functionality, from speech recognition to history storage, works without an internet connection, making it reliable for use on airplanes, in areas with poor connectivity, or for users with strict security requirements.
5. Minimal Resource Footprint: As a native Swift application, Yap is highly efficient. The download is approximately 4 MB, and it idles using around 60 MB of RAM, ensuring it runs smoothly in the background without impacting system performance on Apple Silicon or Intel Macs.

Problems Solved

1. Pain Point: Privacy concerns with cloud-based dictation. Many speech-to-text services send audio to remote servers, creating data privacy and security risks. Yap solves this by guaranteeing audio never leaves the device.
2. Pain Point: Inefficient workflow for frequent dictation. Switching between dedicated dictation apps or web interfaces disrupts workflow. Yap's system-wide hotkey allows instant dictation into any active window.
3. Target Audience: Developers & Technical Writers who dictate code comments, documentation, or prompts for AI coding assistants. Privacy-Conscious Professionals in legal, healthcare, or journalism who handle sensitive information. Productivity-Focused Users who write emails, Slack messages, or notes throughout the day.
4. Use Cases: Dictating complex prompts for AI coding agents (e.g., GitHub Copilot) directly into an IDE; composing confidential client emails or reports; quickly capturing ideas or notes in any application without switching context; writing in environments with no internet access.

Unique Advantages

1. Differentiation vs. Cloud Services: Unlike Google Docs voice typing or Otter.ai, Yap requires no account, no API key, and transmits zero data. It offers lower latency for the initial transcription as there's no network round-trip.
2. Differentiation vs. Built-in macOS Dictation: While macOS has a built-in dictation feature, Yap provides a more streamlined, always-available experience with a dedicated hotkey and a local, searchable history log, which the native tool lacks.
3. Key Innovation: The combination of system-wide input field targeting with fully local, offline speech recognition in a minimal, open-source package. Its ability to paste transcribed text directly into any text field (including terminals, which often block system dictation) is a significant technical and usability advantage.
4. Business Model Transparency: Being free and MIT-licensed eliminates cost barriers and allows for security auditing, building trust within the developer and privacy communities.

Frequently Asked Questions (FAQ)

1. Is Yap really free and what is the license? Yes, Yap is completely free to use and is released under the permissive MIT open-source license, allowing anyone to use, modify, and distribute the software.
2. What are the system requirements for Yap? Yap requires macOS 26 (codenamed Tahoe) or later and runs natively on both Apple Silicon (M-series) and Intel-based Macs. It leverages the Speech framework APIs introduced in this macOS version.
3. How does Yap's accuracy compare to cloud-based AI like OpenAI Whisper? Yap uses Apple's on-device Speech framework, which is optimized for general dictation and balances accuracy with privacy and speed. For highly specialized vocabulary or noisy environments, cloud models may have an edge, but Yap excels in privacy, instant availability, and offline use.
4. Can I use Yap to transcribe pre-recorded audio files? No, Yap is designed specifically for real-time, live dictation. It captures audio directly from your microphone via a hotkey and transcribes it on the fly. It does not support uploading or processing existing audio files.
5. Where is my dictation history stored and is it secure? All transcript history is stored locally in a database on your Mac's storage. It is not encrypted by the app itself but benefits from your system's file security. Since no data is synced to the cloud, it is only accessible on your machine.