BasicGPT: Page & Doc Analyzer, CSV, TTS & Podcast

Summarize web pages, chat with PDF/DOCX, analyze CSV, & generate voice or podcast audio locally. Private WebGPU AI, no cloud.

# BasicGPT: Private AI Suite for Web, Documents, Data & Voice ## Short Description Local AI for pages, PDF, DOCX, CSV, voice, and podcasts. Generate editable podcast scripts, customize host voices, and render/download audio fully on-device. ## Full Description BasicGPT is a privacy-first, local AI browser assistant for Chrome. Analyze web pages, professional documents, and CSV files, convert text into speech, and generate polished podcast scripts with custom host voices, all running entirely on your device with WebGPU acceleration. No cloud APIs, no tracking, and no data leakage. ### 🚀 Six Powerful Workflows **1. PageAI** * **Instant Insights:** Summarize web pages and extract key points in one click. * **Contextual Chat:** Chat directly with the page content for deeper Q&A. **2. DocumentAI** * **Native Support:** Upload PDF, DOCX, and Markdown files. * **Knowledge Retrieval:** Query complex reports and legal papers with precision. **3. PodcastAI ⭐ *NEW*** * **Script-First Generation:** Turn pages or documents into structured podcast scripts. * **Edit Before Render:** Modify lines directly in the generated script before creating audio. * **Custom Host Voices:** Assign different voices to Host A and Host B with built-in conflict protection. * **Portable Workflow:** Import/export podcast script JSON and render/download WAV anytime. **4. VoiceAI** * **Text-to-Speech Studio:** Convert documents (PDF, DOCX, MD) or typed text into high-quality audio. * **One-Click Read from Analysis:** Launch VoiceAI directly from PageAI page info and summary results. * **Streaming + WAV Export:** Start playback quickly with chunked synthesis and create downloadable WAV output. * **Tab-Aware Session State:** Voice text and uploaded document context persist per tab for smoother switching. * **Private Synthesis:** All voice generation happens locally—your voice data never leaves your machine. **5. RowAI** * **Batch Processing:** Run AI actions across CSV rows and export the results. **6. PromptAI** * **Template Library:** Design and save custom AI prompts to reuse across all workflows. ### 🧠 Curated Model Library * **Llama 3.2 3B (Fast & Smart):** Default model for speed/quality balance. * **Qwen 2.5 7B (Pro Max):** Highest quality for complex reasoning. * **Local TTS Engine:** High-fidelity voice synthesis optimized for your device. ### ✨ Key Features * **Multi-Format Export:** Copy results, export to Markdown/Obsidian, or download as **Audio**. * **Right-Click Integration:** Send any selected text directly to PageAI or VoiceAI. * **Model Controls in Settings:** Keep VoiceAI model warm for faster startup or delete cached model files. * **Editable Podcast Scripts:** Fine-tune podcast lines before rendering for better quality and control. * **Per-Host Voice Personalization:** Select distinct voices for each podcast host. * **Podcast Script Portability:** Download and import script JSON for iteration and reuse. * **Performance Mode:** Optimized for WebGPU to ensure smooth inference even on lighter devices. * **Offline Capable:** Works anywhere once the initial models are downloaded. ### 🔒 Privacy & Security * **100% Local Inference:** No data is sent to external servers for analysis or voice synthesis. * **No Telemetry:** We don't track what you read, what you write, or what you listen to. --- ## 🆕 v3.7.0 Release Highlights: PodcastAI Studio Update 🎙️ **PodcastAI Core Experience** * **Create Podcast from Existing Workflows:** Launch PodcastAI directly from page and summary contexts. * **Editable Script Workflow:** Edit generated podcast lines before rendering audio. * **Host Voice Customization:** Configure separate voices for Host A and Host B. * **Integrated Playback Controls:** Unified Play + Settings control with inline voice options. * **Script Portability:** Import/export podcast script JSON for repeatable production. 🧩 **Podcast Quality & Flow Improvements** * **Better Conversational Structure:** Improved planning and refinement for more natural host dialogue. * **Long-Document Coverage:** Chapterized generation path for large inputs with final master-script assembly. * **Stable Render Pipeline:** Improved progress and rendering flow from script to downloadable WAV. --- ## 🆕 v3.6.0 Release Highlights: VoiceAI Maturity Update 🚀 **VoiceAI Core Experience** * **Document-to-Audio:** Transform PDF, DOCX, and Markdown files into listenable content. * **Built-in Playback:** New integrated audio player with speed controls. * **Downloadable Speech:** Save your AI-generated audio for your phone or car. ⚙️ **VoiceAI Reliability & UX Enhancements** * **Offscreen Worker Pipeline:** Non-blocking synthesis path for more stable long-form generation. * **Live Chunk Progress:** Visible synthesis progress during playback and WAV generation. * **Read Anywhere in Workflow:** “Read with VoiceAI” actions added to page details and analysis summary. * **Voice State Persistence:** Voice text and uploaded-doc metadata now persist safely with tab-aware state. * **Voice Model Management:** Configure Keep Warm and clear voice model cache from Settings. 📄 **DocumentAI in v3.6.0** * Native support for professional PDF and Word (.docx) files. * Deep-document chat for research papers and legal analysis. --- **Analyze privately. Edit precisely. Publish podcast-ready audio locally.** --- Checkout the blogs here: https://basicgpt.blogspot.com/2026/03/beyond-cloud-private-ai-with-basicgpt.html