This workflow takes in a PDF file of an installation manual, scans it and then creates FAQs using a GPT4 prompt that will avoid creating questions based on Table of Contents and other rather meaningless content you would not want to be included in a chatbot's knowledge base. The resulting Google sheet can be used in https://gooey.ai/copilot and will provide citations (with page numbers) that link back to relevant section of the source PDF.

Perfect for making FAQs from a collection PDF, webpage and videos to create your own chatbot or enterprise copilot (ie. just add the finished Google sheet url to the doc section in https://gooey.ai/copilot).

In this bulk run example, we compare several different Gooey.AI speech recognition + translations workflows (https://gooey.ai/speech), each of which uses a different AI speech model:

1. Google Chirp
2. Azure
3. Whisper v3
4. Seamless M4T

We then evaluate them at https://gooey.ai/eval/ (where you can see what works best....)

Transcribe any YouTube, mp3, WhatsApp audio or wavs with the best of transcription and translation AI models from OpenAI (Whisper v2 & v3), Microsoft Azure, Google USM, Meta Seemless4MT, AI4Bharat, Bhashini, etc Optionally translate to any language too.

Transcribe mp3, WhatsApp audio + wavs with OpenAI's Whisper or AI4Bharat / Bhashini ASR models. Optionally translate to any language too.