I’m looking for a developer who can build an end-to-end solution that turns Polish call-center recordings into actionable insight. The application may be web-based or desktop—whatever you feel delivers the smoothest workflow—so long as it lets me drag-and-drop audio files and quickly see results. Core requirements • Automatic Polish transcription powered by Google Speech-to-Text • NLP processing (via the OpenAI GPT-4/5 API or a comparable model) that pinpoints key topics and issues—the area I care about most • Summaries that surface recurring complaints, positive feedback, and indicators of agent performance • Simple, intuitive dashboard to upload calls, display transcripts, highlight topics, and show insight cards at a glance • Export options for every report in CSV, JSON, and PDF • Clean, well-commented source code plus step-by-step installation notes I’m open on framework and stack as long as the final build runs reliably on modern Windows and major browsers, and you choose libraries that won’t lock me into expensive licenses. Deliverables 1. Working application (web or desktop) with the features above 2. Source code repository with README and setup script 3. Short demo video or screenshots proving full workflow If you’ve tackled speech analytics, Whisper, Deepgram, or Google Speech projects before, tell me about it—and include a rough timeline.