Yappable for Lovable: voice mode that reads chat & talks back

Every time Lovable finishes a response, Yappable reads it back to you — out loud, in your language, in the level of detail you choose. So you can keep testing, sketching, reviewing, or just thinking, while the agent works in the background. ───────────────────────────────────────────── 🔴 THE PROBLEM EVERY LOVABLE BUILDER KNOWS You send a prompt. You switch tabs to test something. Or you start sketching the next screen. Or you just look away for a second. Then you come back — and now you have to re-read the whole response just to know if it worked. Was that a success? A partial fix? Did it touch something it shouldn't have? Is there a decision waiting for you? That little context switch happens dozens of times a session. It adds up. And no generic text-to-speech tool helps — they just dump raw text at you, full of markdown noise, code paths, and filler, with no sense of what matters. Yappable is built for Lovable specifically. It knows when Lovable is done. It knows what a Lovable response looks like. And it speaks to you the way a co-pilot would — not like a robot reading a file. ───────────────────────────────────────────── 🎛️ FOUR NARRATION MODES — PICK YOUR DEPTH ⚡ FAST: For builders who just need the verdict. Gets straight to the point: is there a pending decision, or are you clear to continue? Nothing else. Best when you're moving quickly and trust the agent. 🌱 BEGINNER: The default. Perfect if you're still learning Lovable. Plain-language summary of what happened — no jargon, no technical terms. You hear what changed, what it means for your project, and what to check. Like having a patient co-builder explaining every move. 🛠️ ADVANCED: For experienced builders who want the full picture, fast. Technical verdict, real risks, any pending decision — using the actual terms. Schema changes, environment variables, API surface, performance claims. The stuff that matters before you ship. 📖 FULL: When you need every word. Reads the entire Lovable response verbatim. Great for long architectural explanations, debugging sessions, or anything you'd normally read start to finish. Slows down slightly for dense content, reads naturally. ───────────────────────────────────────────── 🧠 WHY YAPPABLE IS IN A DIFFERENT CATEGORY THAN ANY TTS TOOL ✅ Built for Lovable's response format — not a general-purpose reader ✅ Yappable understands what's happening instead of just reading what's on screen. ✅ Speaks in your language — not necessarily the language Lovable replied in ✅ Knows the difference between a success, a partial fix, and an error ✅ Flags the things you actually miss — performance claims, copy edits, schema changes, SEO updates — before you trust the green checkmark ✅ Narrates live while the agent is still working, not just when it's done ✅ Knows when Lovable goes silent — and tells you the real status, not just "in progress" ✅ Works with premium voice (ElevenLabs) or your browser's built-in voice ───────────────────────────────────────────── 🚀 HOW TO GET STARTED — 3 STEPS Step 1 — Install and open Lovable Pin the Yappable icon to your toolbar. Open any project on lovable.dev. The extension activates automatically — no setup required to try it. Step 2 — Configure your preferences (inside the popup) Click the Yappable icon in your toolbar. You'll find: * Your narration mode selector (Fast / Beginner / Advanced / Full) * Language picker — choose the language Yappable speaks in * Voice selector — browser native or ElevenLabs * Optional: paste your ElevenLabs API key for premium voice quality. Verification is instant. Your key never leaves your device. Step 3 — Send a prompt and keep working That's it. The moment Lovable finishes, Yappable speaks. You don't have to click anything, switch tabs, or check in. Just listen. ───────────────────────────────────────────── 📡 LIVE NARRATION — STAY INFORMED WHILE IT'S STILL BUILDING Turn on Verbose Mode to hear Lovable's background task steps in real time — each step narrated as it happens, not just the final result. Great for long builds where you want to stay oriented without watching the screen. The silence monitor also watches for stalls. If Lovable goes quiet mid-task, Yappable speaks up with the real status, elapsed time, and the task name — so you always know what's happening, even when nothing appears to be moving. ───────────────────────────────────────────── 🌍 SPEAKS YOUR LANGUAGE — EVEN WHEN LOVABLE DOESN'T Lovable may switch between English and Portuguese (or any other language) mid-session. Yappable normalizes everything into the language you chose — narration, live progress updates, and alerts. One consistent voice, one consistent language, no matter what the agent replies in. Translation runs on-device via Chrome's built-in APIs. Nothing is sent to an external service for this step. ───────────────────────────────────────────── 🔍 PRE-SHIP CHECKS — YAPPABLE WATCHES FOR: ⚠️ Unvalidated performance claims ("it's now faster" with no proof) ⚠️ UI copy that was silently changed ⚠️ Database schema edits or migrations ⚠️ SEO metadata that was touched ⚠️ Build configuration or environment variable changes You hear a flag when any of these appear — so you know where to look before you move on. ───────────────────────────────────────────── 🎙️ PREMIUM VOICE WITH ELEVENLABS The browser's built-in voice works great. But if you want something that sounds natural — especially during long sessions — Yappable supports ElevenLabs voice generation. Bring your own ElevenLabs API key Paste it once inside the popup — verification is instant Your key is stored in chrome.storage.local, never synced, never shared Only the final narration text is sent to ElevenLabs to generate audio ───────────────────────────────────────────── 🧩 EVERYTHING ELSE ✔️ Animated waveform bar while speaking — visual confirmation it's running ✔️ Repeat button — replay the last narration anytime ✔️ Instant stop — silence it mid-sentence ✔️ Multi-tab awareness…