https://chineseman.net/apps/Recent content in Apps on Chinese ManHugoen-usSun, 21 Jun 2026 00:00:00 +0000https://chineseman.net/building-private-speech-to-text-whisper-app/Sun, 21 Jun 2026 00:00:00 +0000https://chineseman.net/building-private-speech-to-text-whisper-app/<p>Most transcription apps upload your audio to a server to process it. If you are recording anything sensitive, a doctor&rsquo;s visit, an interview, a private voice note, a business call, that is a hard no for me. So I built <strong>Private Transcribe</strong>: speech-to-text powered by <strong>Whisper</strong>, running entirely on your device. The audio never leaves your phone. Here is how it came together, and why one promise shaped every technical decision along the way.</p>https://chineseman.net/building-capybara-crossing-arcade-game/Thu, 18 Jun 2026 00:00:00 +0000https://chineseman.net/building-capybara-crossing-arcade-game/<p>After a long run of utility apps, I wanted to make something that was just fun. Something my niece could pick up in two seconds with no tutorial and no explanation. The &ldquo;hop across endless lanes of traffic&rdquo; formula is a classic for a reason, so I dropped a <strong>capybara</strong> into it and built <strong>Capybara Crossing</strong>: hop across roads, train tracks, and rivers, grab coins, try not to die. It turned out that a casual game is a completely different discipline from a tool, and it taught me things the tools never did.</p>https://chineseman.net/building-offline-ai-chat-app-personal-llm/Sun, 14 Jun 2026 00:00:00 +0000https://chineseman.net/building-offline-ai-chat-app-personal-llm/<p>Every mainstream AI chat app ships your conversations off to someone&rsquo;s server. I wanted the opposite: a model that runs on your phone, messages that never leave the device, no login, no subscription. That app became <strong>Personal LLM</strong>, and getting a language model to behave itself on a phone was the most interesting engineering problem I have taken on alone.</p> <div class="app-shots"> <img src="https://chineseman.net/img/llm/1.png" alt="Personal LLM model list showing Qwen 3 8B, Qwen 3 VL 8B vision, and Qwen 3 4B with sizes and RAM requirements" loading="lazy"> <img src="https://chineseman.net/img/llm/2.png" alt="Personal LLM model list showing Qwen 3 1.7B, GLM-Edge 1.5B, and Qwen 3 0.6B on-device language models" loading="lazy"> <img src="https://chineseman.net/img/llm/3.png" alt="Personal LLM chat screen running Qwen 3 0.6B entirely on-device, replying to a message while offline" loading="lazy"> </div> <h2 id="the-itch"> The itch <a class="anchor" href="#the-itch">#</a> </h2> <p>Three things kept bugging me about the mainstream AI apps I was using.</p>