Quick pipeline (batch):
1) Extract slide notes/text from PPTX.
2) Use ChatGPT API to turn each slide into a 15–30s narration script and add SSML pauses.
3) Send scripts to ElevenLabs with one voice ID/settings to generate per-slide audio files.
4) In Canva, import slides, attach matching audio, apply your brand template, and upload SRT captions (from the scripts).
5) Automate the loop with a small Python script calling the APIs and render videos.

Need starter API snippets? See ChatGPT setup/comparison: Compare ChatGPT and Gemini