PROD-RegenAg-EN-DATA

by

Synthetic data generation for RegenAg En Manual

3mo ago

🤖 Youtube/PDF/Drive/Web URLs

Loading...

Show as Links

🩻 Document Intelligence Model

📊 Google Sheets URL (optional)

⚙️ Settings

👩‍🏫 Task Instructions

You are extracting Q&A pairs from a regenerative farming training manual 
for use in a RAG system. The end users are smallholder farmers in 
sub-Saharan Africa. English may not be their first language, and their 
questions may have been translated from a local language.

WHAT TO EXTRACT FROM:
- Main instructional content about farming practices, soil health, 
  crop management, land preparation, pest management, and related topics

WHAT TO IGNORE:
- Questions that a facilitaor might ask
- Session delivery instructions, ice-breaker suggestions, wrap-up notes
- Table of Contents, module title pages, image captions, icon pages
- Any text that refers the reader to another section for the answer
- Discussion questions posed to farmer participants
- Passing mentions or incomplete ideas
- If the transcript contains no substantive farming content, 
  output nothing at all

OUTPUT FORMAT:
Begin with a single section heading, then list all Q&A pairs for 
this page beneath it.

sectionHeading=Generated FAQ
content=
For each extractable concept:
- Write 2 to 3 differently phrased questions that lead to the same 
  answer. Phrase them as a smallholder farmer would naturally ask — 
  short, simple, practical English. Vary the phrasing.
- Write one clear, complete answer in plain simple English. 
  Short sentences. If a technical term is needed, explain it 
  briefly inline (e.g. "mulch — a layer of plant material 
  covering the soil").

Format each block as:
Question one phrasing?
Question two phrasing?
Question three phrasing?
Answer text here.

Separate each Q&A block with a blank line.
No numbering, bullet points, headers, or commentary beyond the format above.
If there is no valid content to extract, output nothing at all.

🔠 Language Model Settings

Avoid Repetition

Response Format

Max Output Tokens

Creativity (aka Sampling Temperature)

Answer Outputs

Reasoning Effort

🦻 Speech Recognition & Translation

Recognize speech and translate for audio and video files.

Filter by Language

Speech-to-Text Provider

Spoken Language

🔠 Translate

🛠️ Developer Tools and Functions

Run cost = 500 credits

With each run, you agree to Gooey.AI's terms & privacy policy.

Input Documents

🔗https://drive.google.com/file/d/1PihkSLvBrrp-QV0Oq7MKC4AqcFrnwqHR/view?usp=drivesdk

🔗https://drive.google.com/file/d/16p9P5GVEL6xfXMYlUmiGmCDMmH89KpxG/view?usp=drivesdk

🔗https://drive.google.com/file/d/1EObxS21AL-EUbbz5lCemAWsGjxglBDO8/view?usp=drivesdk

🔗https://drive.google.com/file/d/1AiLxv6rl0T9Y4gjQ_Z8JTv2B0OzUXscb/view?usp=drivesdk

🔗https://drive.google.com/file/d/1Zu49xJ8tAGDTnT5Mj6-Wlzu0wzRyUBLq/view?usp=drivesdk

🔗https://drive.google.com/file/d/1MezDJ3fYmMhitlvYvkNQ8aR19EcRbm68/view?usp=drivesdk

🔗https://drive.google.com/file/d/1hHZ56aTKZJIYym5R-d5C0Bc40fQw1vwb/view?usp=drivesdk

🔗https://drive.google.com/file/d/1b78kwHlKbBrlLAkzjjmV5LVWZA8rr4KY/view?usp=drivesdk

🔗https://drive.google.com/file/d/1saCvm_QMH8tiK68U6FmUJs8MyXBgzn2L/view?usp=drivesdk

🔗https://drive.google.com/file/d/1K-B8CnVNgo3kx8K_4R7HB6hO1Z4SwDrk/view?usp=drivesdk

🔗https://drive.google.com/file/d/1rgYyhvGAlV0hgw2EVjOQNAYjlkZZSz6T/view?usp=drivesdk

🔗https://drive.google.com/file/d/13GcQwEHODgNUr7Wq2yV7aqNDq9wecRV2/view?usp=drivesdk

🔗https://drive.google.com/file/d/1IDCp26cf6emTr0lZqBrMZ6d0tL4A_v2e/view?usp=drivesdk

🔗https://drive.google.com/file/d/1oIT4pAgixd98cZKF4WeU9WxFZfYom4EK/view?usp=drivesdk

🔗https://drive.google.com/file/d/13mNL4aDTYNazhoSxnV0BRxEqHXq2UbhR/view?usp=drivesdk

🔗https://drive.google.com/file/d/1vdQClJMArFHOZoXoFfqekFc7-rgg80Ft/view?usp=drivesdk

🔗https://drive.google.com/file/d/1P5oqGK6C2uSz4RzxyhLEgwjFqjkv5-EQ/view?usp=drivesdk

🔗https://drive.google.com/file/d/15ksEHqA54DlyfzQjtxtoIEColPHxE-LD/view?usp=drivesdk

🔗https://drive.google.com/file/d/1v5KOUExpOFpmiYCnWSMC3eGDpVWi2I1v/view?usp=drivesdk

🔗https://drive.google.com/file/d/1QYSAp9VfkjqsKe51aYYMFTxz-TJBOW6S/view?usp=drivesdk

🔗https://drive.google.com/file/d/1iZ7PeCP1TbJPBdsn6T1wgFmFsR-Ntksz/view?usp=drivesdk

🔗https://drive.google.com/file/d/1G4xNYs2RcspM_A7sRkVCiwD9BgTONx3i/view?usp=drivesdk

🔗https://drive.google.com/file/d/1FpuWsh5LO1cDKT41QuFoPEyGXhSbvvWX/view?usp=drivesdk

🔗https://drive.google.com/file/d/1-bNSB0TR9nGJAZPvgAkgJLzOXCrFpgJx/view?usp=drivesdk

🔗https://drive.google.com/file/d/19hGnoA1qQOgHmrm8pAjVFi1fxmmd_dym/view?usp=drivesdk

🔗https://drive.google.com/file/d/1MAVMon2JQ10FF3RHZcX57Wc0LPKSdq4l/view?usp=drivesdk

🔗https://drive.google.com/file/d/1zagxydHw7J4WjOMOKZo-RNWUuGUjp-Hc/view?usp=drivesdk

🔗https://drive.google.com/file/d/1ylgsWdGRQn1WF7_ZrNuffFW6949xIh0x/view?usp=drivesdk

🔗https://drive.google.com/file/d/1gLVjbweGr01QTpOgTY8mSJ4VVGFQPxVl/view?usp=drivesdk

🔗https://drive.google.com/file/d/1O0kn3yO1uI5n0poa3NNtkcVbDnS-anK5/view?usp=drivesdk

🔗https://drive.google.com/file/d/1SupJJOkd8kWEpiuU4xEYZiHbvueTLA4x/view?usp=drivesdk

🔗https://drive.google.com/file/d/1p1sveQ80j576sSvTD9KVZbGh9seOmR7z/view?usp=drivesdk

🔗https://drive.google.com/file/d/1KSRTr9T8Cmdk45xx0plVmodb42w7XSVq/view?usp=drivesdk

Google Sheets URL

https://docs.google.com/spreadsheets/d/1X0NCO4Iq1eYLizJtLfXyc1w7-z8AM0xJdKfA22XajlE/edit?usp=sharing

🐞 Debug

How to Use This Recipe

🙋🏽‍♀️ Need more help? Join our Discord

Related Workflows

Agent Builder

Gooey.AI's base AI workflow with built-in RAG, web search, voice understanding of 1000+ languages, code creation + execution, API connections & integrations to create your own WhatsApp, Web, FB and voice AI …

Speech Recognition and Translation

Transcribe mp3s, WhatsApp voice, YouTube videos in 1000+ langs with Meta’s MMS /Seemless M4T, OpenAI's GPT4o Audio LLM, Whisper v2/v3, Azure, Google, GhanaNLP, AI4Bharat & Bhasini ASR models. Optionally …

Compare LLMs: GPT5.4, Deepseek3.2, Claude4.6, Gemini3.1, Apertus, Kimi 2.5 vs GLM5

Which language model works best for your prompt? What are the biases inherent in each? Compare the latest models from Google, Deepseek, OpenAI, Mistral, Meta, SwissAI OpenAI and Anthropic with more LLMs being …

RAG in the Cloud: Search any document with AI

We've built the best Retrieval Augmented Generation (RAG) as-a-Service anywhere - now with page-level citations! Absorb tables, PDFs, docs, links, videos or audio clips and use our synthetic data maker to …