Home on ApplauseLab

API Reference

Mon, 01 Jan 0001 00:00:00 +0000

Yap exposes Go methods to the React frontend via Wails bindings. This reference documents all available methods.

State Management

GetState

Get the current application state.

import { GetState } from '../wailsjs/go/main/App';

const state = await GetState();

Returns: AppState

interface AppState {
 state: 'ready' | 'recording' | 'transcribing' | 'error';
 recordingTime: number; // Duration in seconds
 lastTranscript: string; // Most recent transcription
 error: string; // Error message if state is 'error'
 currentModel: string; // Active model name
 currentProvider: string; // 'local' or 'openai'
 modelReady: boolean; // Whether model is ready
 hotkeyEnabled: boolean; // Whether hotkey is active
}

Recording

ToggleRecording

Start or stop recording based on current state.

Building from Source

Mon, 01 Jan 0001 00:00:00 +0000

Prerequisites

Before building, ensure you have the following installed:

Required

Tool	Version	Installation
Go	1.21+	golang.org
Node.js	18+	nodejs.org
Wails CLI	v2	`go install github.com/wailsapp/wails/v2/cmd/wails@latest`
PortAudio	Latest	See below

Optional (for local transcription)

Tool	Purpose	Installation
whisper-cpp	Local transcription	`brew install whisper-cpp`

Installing PortAudio

macOS

brew install portaudio

Ubuntu/Debian

sudo apt-get install portaudio19-dev

Fedora

sudo dnf install portaudio-devel

Windows

PortAudio is included via CGO bindings. Ensure you have a working C compiler (MinGW or MSVC).

Installation

Mon, 01 Jan 0001 00:00:00 +0000

Download

Choose the installer for your platform:

Platform	Download	Notes
macOS	DMG Installer	Apple Silicon native, Intel via Rosetta
Windows	Installer	Windows 10/11 (x64)
Linux	AppImage / DEB	Ubuntu, Debian, and most distros

See all versions on the Releases page.

macOS Installation

Download the DMG file from the releases page
Open the DMG file
Drag Yap to your Applications folder
Open Yap from Applications
If you see a security warning, go to System Settings → Privacy & Security and click “Open Anyway”

Windows Installation

Download the installer (.exe) from the releases page
Run the installer
Follow the installation wizard
Launch Yap from the Start Menu

Linux Installation

AppImage

# Download the AppImage
chmod +x Yap-*.AppImage
./Yap-*.AppImage

Debian/Ubuntu (.deb)

sudo dpkg -i Yap-*.deb

System Requirements

macOS: 10.15 (Catalina) or later
Windows: Windows 10/11 (x64)
Linux: Modern distribution with glibc 2.31+
RAM: 2GB minimum, 4GB+ recommended for larger models
Storage: 100MB for app, plus space for whisper models (75MB - 3GB)

Next Steps

After installation, you’ll need to:

Recording

Mon, 01 Jan 0001 00:00:00 +0000

Starting a Recording

There are two ways to start recording:

Using the Global Hotkey

Press your configured hotkey (default: Right Option ⌥) from any application. You’ll see:

A visual overlay indicating recording is active
An audio cue (if enabled)

Using the App Interface

Open the Yap window
Click the Start Recording button

Stopping a Recording

Press the hotkey again or click Stop in the overlay. The recording will be:

Saved temporarily
Sent for transcription
Auto-pasted to your active application (if enabled)

Canceling a Recording

If you want to discard a recording without transcribing:

Configuration

Mon, 01 Jan 0001 00:00:00 +0000

Configuration File

Yap stores settings in a JSON configuration file:

Location:

macOS: ~/Library/Application Support/yap/config.json
Windows: %APPDATA%\yap\config.json
Linux: ~/.config/yap/config.json

Configuration Options

provider

Transcription provider to use.

Value	Description
`"local"`	Use whisper.cpp (default)
`"openai"`	Use OpenAI Whisper API

{
 "provider": "local"
}

model

Active Whisper model for local transcription.

Value	Size	Notes
`"tiny"`	~75 MB	Fastest, multilingual
`"tiny.en"`	~75 MB	Fastest, English-only
`"base"`	~150 MB	Good balance, multilingual
`"base.en"`	~150 MB	Good balance, English-only (default)
`"small"`	~500 MB	Better accuracy, multilingual
`"small.en"`	~500 MB	Better accuracy, English-only
`"medium"`	~1.5 GB	High accuracy, multilingual
`"medium.en"`	~1.5 GB	High accuracy, English-only
`"large-v3"`	~3 GB	Highest accuracy, multilingual

{
 "model": "base.en"
}

openaiApiKey

API key for OpenAI Whisper (required for cloud provider).

Project Structure

Mon, 01 Jan 0001 00:00:00 +0000

Overview

Yap is built with Wails, combining a Go backend with a React frontend.

yap/
├── app.go # Main application logic
├── main.go # Wails entry point
├── internal/ # Go packages
├── frontend/ # React application
└── build/ # Build assets and output

Entry Points

main.go

Wails application entry point. Configures:

Window size and properties
Menu binding
Asset handling

app.go

Core application struct with methods exposed to the frontend:

Recording state management
Transcription coordination
History management
Configuration

Internal Packages

internal/audio

Audio recording using PortAudio.

Quick Start

Mon, 01 Jan 0001 00:00:00 +0000

This guide will have you dictating text in under 2 minutes.

Step 1: Launch Yap

Open Yap from your Applications folder (macOS), Start Menu (Windows), or application launcher (Linux).

You’ll see the main window with:

Home — Recording controls and status
Settings — Configure transcription and hotkeys
History — View past transcriptions

Step 2: Download a Model (Local Mode)

If you’re using local transcription (the default):

Go to Settings
Under Transcription Provider, ensure “Local” is selected
Click on a model to download it

We recommend starting with base.en — it’s a good balance of speed and accuracy.

Transcription

Mon, 01 Jan 0001 00:00:00 +0000

Yap supports two transcription providers: Local (using whisper.cpp) and Cloud (using OpenAI’s Whisper API).

Local Transcription

Local transcription processes audio entirely on your device using whisper.cpp.

Advantages

Private — Audio never leaves your device
Offline — Works without internet connection
Free — No API costs or subscriptions

Disadvantages

Slower — Processing depends on your hardware
Resource usage — Uses CPU/RAM during transcription
Less accurate — May struggle with accents or specialized vocabulary

Requirements

You need whisper-cpp installed on your system:

Troubleshooting

Tue, 18 Aug 2026 00:00:00 +0000

Recording Issues

Hotkey Not Working

Symptoms:

Pressing the hotkey does nothing
No recording overlay appears

Solutions:

Check Accessibility permissions (macOS)
- Go to System Settings → Privacy & Security → Accessibility
- Ensure Yap is enabled
- Try removing and re-adding if already enabled
Try a different hotkey
- Go to Settings → Recording Hotkey
- Select an alternative (Left Option, Fn, or Double-tap)
Check for conflicts
- Other apps may be capturing the hotkey
- Try quitting other apps that use global hotkeys
Restart Yap

Contributing

Mon, 01 Jan 0001 00:00:00 +0000

We welcome contributions to Yap. This guide explains how to get involved.

Ways to Contribute

Bug Reports — Found a bug? Open an issue
Feature Requests — Have an idea? Start a discussion
Code — Fix bugs or implement features
Documentation — Improve docs or add examples
Testing — Test on different platforms

Getting Started

Fork the repository on GitHub
Clone your fork locally
Set up the development environment (see Building)
Create a branch for your changes

git clone https://github.com/YOUR_USERNAME/yap.git
cd yap
git checkout -b feature/your-feature-name

Development Workflow

1. Make Changes

Edit code, add features, fix bugs.

Permissions

Mon, 01 Jan 0001 00:00:00 +0000

Yap requires specific system permissions to function. This guide explains what permissions are needed and how to grant them.

Required Permissions (macOS)

Microphone Access

Why needed: To record your voice for transcription.

How to enable:

Open System Settings (or System Preferences on older macOS)
Go to Privacy & Security → Microphone
Find Yap in the list
Toggle it ON

If Yap isn’t listed, try starting a recording — macOS will prompt you to grant access.

Whisper Models

Mon, 01 Jan 0001 00:00:00 +0000

When using local transcription, you need to download a Whisper model. This guide helps you choose the right one.

Available Models

Model	Size	Speed	Accuracy	Best For
`tiny`	~75 MB	Fastest	Basic	Quick tests, low-power devices
`tiny.en`	~75 MB	Fastest	Basic	English-only, slightly better
`base`	~150 MB	Fast	Good	General multilingual use
`base.en`	~150 MB	Fast	Good	Recommended for English
`small`	~500 MB	Medium	Better	When accuracy matters
`small.en`	~500 MB	Medium	Better	Professional English dictation
`medium`	~1.5 GB	Slow	High	Technical/medical content
`medium.en`	~1.5 GB	Slow	High	High-stakes English content
`large-v3`	~3 GB	Slowest	Highest	Maximum accuracy needed

English vs Multilingual

Models ending in .en are English-only. They’re:

FAQ

Mon, 01 Jan 0001 00:00:00 +0000

General

What is Yap?

Yap is a desktop speech-to-text application. Press a hotkey, speak, and your words are transcribed and pasted into any application.

Is it free?

Yes. Yap is free and open source under the MIT license.

What platforms are supported?

macOS: Apple Silicon (native) and Intel (via Rosetta)
Windows: Windows 10/11 (x64)
Linux: AppImage and DEB packages

Does it work offline?

Yes, when using local transcription with whisper.cpp. Models are downloaded once and work entirely offline.

History

Mon, 01 Jan 0001 00:00:00 +0000

Yap keeps a history of your transcriptions so you can review, copy, and replay past recordings.

Viewing History

Click the History tab in Yap
Browse your past transcriptions

Each history item shows:

Transcribed text — The result of your dictation
Timestamp — When the recording was made
Duration — Length of the recording

Actions

Copy Text

Click the Copy button to copy the transcribed text to your clipboard. Useful when auto-paste didn’t work as expected or you need the text again.

Settings

Mon, 01 Jan 0001 00:00:00 +0000

Access Settings by clicking the Settings tab in Yap.

Transcription Provider

Choose between local and cloud transcription:

Local — Uses whisper.cpp on your device
Cloud (OpenAI) — Uses OpenAI’s Whisper API

See Transcription for details on each provider.

Model Selection (Local Only)

When using local transcription:

Browse available models
Click to download models you need
Click a downloaded model to select it as active

The active model shows a green checkmark.

Keyboard Shortcuts

Mon, 01 Jan 0001 00:00:00 +0000

Global Hotkeys

These work from any application:

Action	Default Shortcut	Notes
Start/Stop Recording	`Right ⌥`	Configurable in Settings
Cancel Recording	`Esc`	During active recording

Alternative Hotkey Options

Configure in Settings → Recording Hotkey:

Option	Key Combination
Right Option	`⌥` (right side)
Left Option	`⌥` (left side)
Fn	`fn`
Double-tap Right Option	`⌥` `⌥` (quick double-tap)

Recording Overlay

When the recording overlay is visible:

Action	Shortcut
Stop & Transcribe	`Return` or hotkey
Cancel Recording	`Esc`

Application Window

When Yap window is focused:

Trifon Getsov

Mon, 15 Jan 2024 00:00:00 +0000

Co-Founder & CEO @ ApplauseLab

January 2025 - Present

Leading a specialized R&D and idea acceleration lab focused on building products and helping founders turn B2C ideas into MVPs and secure funding.

Building next-gen products in personal development and wellbeing
Guiding startups from concept to market-ready products
Built smart. Delivered fast. Designed to solve.

Co-Founder & CEO @ From Farm to Table

October 2024 - Present

Building a cutting-edge manufacturing facility that converts agricultural waste into biodegradable materials, advancing sustainable materials and circular economy solutions.

Kris Nikolov

Mon, 15 Jan 2024 00:00:00 +0000

Co-Founder & CTO @ xHeal.ai

October 2025 - Present | Miami, FL (Remote)

Building the future of preventive healthcare - making it understandable, actionable, and affordable.

Launched platform live in the US with Europe expansion coming soon
Architecting 360° healthcare platform that transforms how people think about their health
Creating technology that catches health issues before they become critical
Building scalable infrastructure to support international growth

Cloud Architect @ EquityMultiple

October 2022 - Present | Remote (Contract)

Kalin Stoev

Mon, 15 Jan 2024 00:00:00 +0000

Full-Stack Architect | AI & Mobile Systems @ xHeal.ai

October 2024 - Present | Miami, FL (Remote)

Architected and shipped production AI-powered healthcare platform transforming reactive healthcare into proactive wellness management through intelligent data unification and personalized recommendations.

Architected entire React Native application from scratch following iOS Human Interface Guidelines and best practices, implementing scalable architecture with Zustand for global state management and modular component
Designed the database schema and AI workflow architecture for consistent data flow
Designed and launched production app to US App Store with 5000+ active users, 99.9% uptime, and 4.8★ average rating
Integrated Apple HealthKit API enabling real-time bidirectional synchronization of 50+ health metrics (heart rate, sleep patterns, activity levels, nutrition data) across iOS ecosystem

Lead AI Software Engineer @ ApplauseLab

October 2024 - Present | Remote

Nikolay Kolibarov

Mon, 15 Jan 2024 00:00:00 +0000

Senior Software Engineer @ Finonex

April 2023 – Present | Bulgaria (Remote)

Designed, developed, and maintained a distributed lead recommendation and scoring system for fintech call centers.

Built reactive lead scoring system using Python, Redis, Lua, and Kafka to process 3M+ leads in real-time across thousands of sales representatives.
Developed call dialog synchronization and A/B testing features for lead conversion optimization.
Migrated deployments from raw YAML to Helm for improved maintainability.
Enhanced CI/CD pipelines with automated integration testing and ephemeral environment provisioning in Kubernetes.
Designed APIs with Express, MariaDB, Redis, and Kafka.
Built desktop and web apps for call center agents and support teams.
Reduced AWS RDS costs through data modeling and query optimization.
Developed monitoring dashboards and alerts in Grafana.

Senior Software Engineer @ Fieldin

July 2022 – April 2023 | Bulgaria (Remote)

Samuil Dimov

Mon, 15 Jan 2024 00:00:00 +0000

UX Lead @ xHeal.ai

November 2025 – Present | Miami, Florida, United States (Remote)

Leading the design of xHeal’s user experience, creating intuitive flows, dashboards, and logging tools that help users manage chronic conditions. Conducting user research and collaborating with product teams.

Designer @ The Missing Link Tech

December 2023 – November 2025 | Sofia, Bulgaria (Hybrid)

Full-time designer focusing on web design, UI testing, and creating user-centered digital experiences.

Workflow catalog

Fri, 07 Aug 2026 00:00:00 +0000

Hamal: An Agentic Harness for Local Repository Work

Mon, 03 Aug 2026 00:00:00 +0000

Most coding agents look straightforward while they are still a prompt box. The difficult part starts when a model asks to read a file, write code, run a command, pause for input, or continue later without losing what happened.

Hamal is our attempt to make that difficult part explicit.

It is an experimental, Effect-native local coding agent organized as an npm workspace. Effect is the typed runtime and composition foundation beneath the workspace, not a wrapper added at the edge.

The product navigation calls Hamal an agentic harness. That means a bounded model and tool-execution loop around local repository work, not an autonomous magic box.

TL;DR: Hamal keeps project selection, provider and model choice, tools, permissions, durable run state, context management, and capability UI inside one local harness. It supports an interactive terminal UI and a plain headless mode. It is currently an experimental private preview, its APIs can change, no open-source license has been selected, and it is not ready for redistribution.

Introducing xHeal V2: One App for Your Health, Powered by ApplauseLab

Sun, 12 Jul 2026 18:00:00 +0300

Your health data should not be scattered across apps, PDFs, wearables, and notes.

xHeal V2 brings health records, lab results, Apple Health data, workouts, nutrition, mindfulness, and daily health metrics into one secure app. It helps people organize their information, understand wellness patterns, and build healthier habits with AI-powered insights grounded in the information they choose to add.

V2 is more than a visual update. It is the clearest expression yet of what we are building at ApplauseLab: a connected product powered by our reusable Infrastructure Layer and Intelligence Layer.

xHeal is the flagship venture and primary product of our venture studio. It is where those layers are being tested, refined, and proven in a real product people can download and use today.

From Digithon to Digitalk: Bulgaria's Tech Ecosystem Has Never Felt More Alive

Sun, 12 Jul 2026 17:30:00 +0300

What were you building at 17 or 18 years old?

At Digitalk Conference in Sofia, I watched three teams of founders aged 17-19 present products live in front of a jury and the full conference audience.

They were the three teams that had earned the top places at Digithon two days earlier. Now their ideas had moved from a fast-paced hackathon into one of Southeast Europe’s leading technology and business events.

The products were impressive. The bigger story was the ecosystem forming around the people who built them.

Leading by Example: Mentoring the Next Generation at Digithon

Sun, 12 Jul 2026 00:00:00 +0000

Four students. One ambitious idea. Just a couple of hours to turn it into a working MVP.

On June 7, I joined Digithon in Sofia as a mentor to a team of four students from TUES. They were in 10th and 11th grade, aged 16-17, but they approached the challenge with the energy and ambition of a team preparing to launch a real product.

By the end of the day, they had built an AI-assisted revenue management system designed for large grocery chains in the United States.

That result was impressive. The people behind it were even more impressive.

Autonomous Development System

Sun, 05 Jul 2026 00:00:00 +0000

Bachkator: Terraform for Your Agent Loops

Wed, 10 Jun 2026 00:00:00 +0300

Its name comes from Bulgarian: бачкатор (pronounced BAHCH-kah-tor) means “the hard worker,” the one who gets the job done.

We have been using the tool internally for about a month, and it has been useful enough that we decided to clean it up and share it.

Temporal Workflows, Now One Raycast Command Away

Sun, 07 Jun 2026 00:00:00 +0000

We use Temporal every day at Applause Lab. It is how we build reliable, long-running systems without turning every retry, timeout, and recovery path into application glue.

But when you live in Temporal daily, the small operational loops start to add up. You jump between the Temporal UI, the CLI, local namespaces, staging, production, workflow histories, schedules, and a browser tab you probably already lost under five others.

Raycast is already where we open tools, run commands, and move through our work. So we brought Temporal into that flow.

Yap: A Lightweight, Open-Source Dictation Tool

Mon, 01 Jun 2026 00:00:00 +0000

We built Yap because we wanted a dictation tool that felt like a utility, not a platform.

Lightweight. Minimal. Open source. Easy to run. Easy to understand. Easy to change if your workflow needs something different.

The basic loop is simple: press a hotkey, speak, transcribe, and paste the result into whatever app already has focus. No new writing surface. No heavy workspace. No assumption that everyone wants the same workflow.

Replatforming EquityMultiple from Heroku to AWS

Fri, 22 May 2026 00:00:00 +0000

Operating Intelligence Platform for Customer Success

Wed, 13 May 2026 00:00:00 +0000

Bakery Operations Platform & E-Commerce Backbone

Fri, 01 May 2026 00:00:00 +0000

AI Career Platform Built from Idea to POC

Wed, 15 Apr 2026 00:00:00 +0000

AI-Enablement for Engineering & Development Teams

Wed, 01 Apr 2026 00:00:00 +0000

Full-Stack Technical Consulting for AI E-Commerce Platform

Sun, 15 Mar 2026 00:00:00 +0000

Visitor Management System for National Security Facility

Sun, 01 Mar 2026 00:00:00 +0000

Introducing Atelier and the ServiceNow Architect AgentPack

Mon, 12 May 2025 00:00:00 +0000

The workflows that run a company are not neat little tasks. They live across email, documents, delivery notes, project history, approvals, and the judgement of the people doing the work.

That is what we built Atelier for: hyperautomation for the work that actually moves the business.

Atelier ingests operational data, encodes an AgentPack against how a business actually decides, and runs it alongside the team. HR, delivery, operations, marketing, engineering, the workflows that carry the company can become runnable systems instead of scattered process knowledge.

About ApplauseLab

Mon, 01 Jan 2024 00:00:00 +0000

Download Yap

Mon, 01 Jan 2024 00:00:00 +0000

Atelier, the Operating Core for Supervised AI Work

Mon, 01 Jan 0001 00:00:00 +0000

Privacy Policy

Mon, 01 Jan 0001 00:00:00 +0000

Last updated: August 7, 2026

This policy explains how ApplauseLab handles personal data when you visit applauselab.ai, contact us, or book a workflow call. It does not cover customer projects, which are governed by their own agreements and data-processing terms.

Data we collect

We may collect:

Information you send us, including your name, email address, company, message, and booking details.
Basic technical records created when the site is served, such as IP address, browser and device type, requested pages, timestamps, referrer, and security logs.
Calendar data you provide to the external booking provider after choosing to open or use its calendar.

We do not offer website accounts or run advertising trackers on this site.

Speak with an AI Architect

Mon, 01 Jan 0001 00:00:00 +0000

Terms of Service

Mon, 01 Jan 0001 00:00:00 +0000

Last updated: August 7, 2026

These terms govern your use of applauselab.ai. Customer projects, subscriptions, and other paid work are governed by the agreement made for that work. If a project agreement conflicts with these website terms, the project agreement controls.

Using this website

You may use the website for lawful purposes. You may not attempt to disrupt the site, bypass its security, misuse its content, or use it in a way that infringes another person’s rights.