Add blog post: Moonshine Micro runs speech recognition and TTS in 470 KB on an 80-cent chip

Author: cappy-dev

blog/index.html

@@ -46,6 +46,22 @@ <h1>Blog</h1>
 
         <div class="post-month" id="m-july">July</div>
 
+        <a class="post-row" data-topics="moonshine micro speech-recognition tts edge-ai rp2350 mcu whisper local-ai tflm hacker-news moonshine-voice useful-sensors open-source" href="/blog/moonshine-micro-500kb-speech-mcu.html">
+            <div class="post-date">Jul 19<span class="fresh" data-published="2026-07-19"><span class="fresh-dot"></span>new</span></div>
+            <div class="post-content">
+                <h2 class="post-title">Moonshine Micro runs speech recognition and TTS in 470 KB on an 80-cent chip</h2>
+                <p class="post-excerpt">Moonshine Micro, from the Useful Sensors team behind the larger Moonshine Voice toolkit, runs voice activity detection, speech-to-text, and neural text-to-speech in 468 KB of RAM on a Raspberry Pi RP2350 that costs about 80 cents. The sub-500 KB number is real. The catch the HN headline glosses over is that the micro build's STT is a spelling-based command recognizer, not the free-form Whisper-class transcriber that the bigger Moonshine Streaming models provide. Both halves of the project, what each one actually does, and which one to reach for.</p>
+                <div class="post-tags">
+                    <span class="pill">Moonshine</span>
+                    <span class="pill">Speech Recognition</span>
+                    <span class="pill">Edge AI</span>
+                    <span class="pill">RP2350</span>
+                    <span class="pill">Local AI</span>
+                    <span class="pill">Hacker News</span>
+                </div>
+            </div>
+        </a>
+
         <a class="post-row" data-topics="gpt-5.6 openai convex-optimization lower-bound lipschitz proof llm-research hacker-news math" href="/blog/gpt-5-6-convex-optimization-proof.html">
             <div class="post-date">Jul 19<span class="fresh" data-published="2026-07-19"><span class="fresh-dot"></span>new</span></div>
             <div class="post-content">