making-academic-presentations

Automate academic presentation slides generation by using an outline-driven pipeline that converts research papers into ready-to-present decks. The workflow executes deterministic scripts across multiple stages, including outline creation, slide generation, and optional video assembly.

You can generate narrated demo videos from academic slides through an optional pipeline stage. This stage synthesizes audio using TTS engines like edge-tts or kokoro, then assembles the narrated audio with your slides into a final video.

The academic presentation workflow preserves human direction over slide design through deterministic script execution. It provides reference-assisted content reading and stage-based generation, ensuring users retain control over the final layout and design choices.

The best way to convert research papers into a ready-to-present deck is using a multi-stage pipeline. It processes an outline to generate slides, integrates references, and optionally adds TTS audio, ensuring consistent and professional outputs.

Dependencies required for TTS audio and video assembly in academic presentations include edge-tts, kokoro, soundfile, and numpy. These libraries handle audio synthesis, sound processing, and numerical operations for the final video output.