Gooey.AI

Neo-natal AI Advisor for Indian Nurses

An interactive AI bot that guides rural nurses in India with advice for pregnant mothers based on a PDF of the WHO guidelines. A collaboration between Gooey.AI and IPRD Solutions.

📝 Instructions

You are an intelligent AI assistant. Follow the instructions as closely as possible. If the user wants to reset the chat, print <|reset|>.

health worker: Let's role play a clinical situation. I'll play the role of a community health worker at a patient's home in Gujarat and you can be a nurse on phone in a rural Public Health Center in Gujarat. You tend to be concerned with identifying high risk illnesses in patients such as preeclampsia and malaria.  You are helping me attend to the patient at her home in my village. I am skilled to use thermometer, blood pressure meter, SPO2 device and stethoscope. I can collect urine samples. I have the following supplies: Paracetamol tablets, IpTpSp tablets, Anti-Tetanus shots, and malaria RDTs. Please generate your answers based primarily. Use an unbiased, succinct, and friendly tone. Think step-by-step and ask questions one at a time. Respond in the same language as the last question or statement uttered by the Health worker. Do not quote the sources and search results in your responses. Highlight actions and recommendations as seperate itemized lists in your responses. Please complete the last sentence even if it does not fit into maximum number of token.

nurse: Sure, let's get started! I'm ready to assist you in attending to the patient. What seems to be the patient's concern today?

🧠 Language Model

Mixtral 8x7b Instruct v0.1 (Mistral)

📄 Knowledge

Submit Links in Bulk

💪 Capabilities

🦻 Speech Recognition & Translation

🗣️ Text to Speech & Lipsync

Text-to-Speech Provider

Google Text-to-Speech

Voice name (Google TTS)

en-GB-Neural2-F (Female)

Please refer to the list of voice names here

🫦 Add Lipsync Video

👩‍🦰 Input Face

Upload a video or image (with a human face) to lipsync responses. mp4, mov, png, jpg or gif preferred.

Lipsync Model

SD, Low-res (~480p), Fast (Rudrabha/Wav2Lip)

🩻 Photo & Document Intelligence

🧩 Developer Tools and Functions

⚙️ Settings

🗣️ Google Text-to-Speech Settings

Speaking rate

1.0 is the normal native speed of the speaker

Pitch

Increase/Decrease semitones from the original pitch

⌖ Lipsync Face Padding

Adjust the detected face bounding box. Often leads to improved results. Recommended to give at least 10 padding for the chin region.

Head

Chin

Left Cheek

Right Cheek

📄 Knowledge Base

👩‍🏫 Search Instructions

How should the LLM interpret the results from your knowledge base?

Citation Style

Numbers ( [1] [2] [3] ..)

🔗 Shorten Citation URLs

Create Synthetic Data

To improve answer quality, pick a synthetic data maker workflow to scan & OCR any images in your documents or transcribe & translate any videos. It also can synthesize a helpful FAQ. Adds ~2 minutes of one-time processing per file.

———

Advanced Settings

In general, you should not need to adjust these.

👁‍🗨 Conversation Summarization

These instructions run before the knowledge base is search and should reduce the conversation into a search query most relevant to the user's last message.

🔑 Keyword Extraction

Instructions to create a query for keyword/hybrid BM25 search. Runs after the Conversations Summarization above and can use its result via {{ final_search_query }}.

✏ Embeddings Model

Text Embedding 3 Large (OpenAI)

Dense Embeddings Weightage

Weightage for dense vs sparse embeddings. 0 for sparse, 1 for dense, 0.5 for equal weight.
Generally speaking, dense embeddings excel at understanding the context of the query, whereas sparse vectors excel at keyword matches.

Max Citations

The maximum number of document search citations.

Max Snippet Words

After a document search, relevant snippets of your documents are returned as results.
This setting adjusts the maximum number of words in each snippet (tokens = words * 2).
A high snippet size allows the LLM to access more information from your document results, at the cost of being verbose and potentially exhausting input tokens (which can cause a failure of the copilot to respond).

Snippet Overlap Ratio

Your knowledge base documents are split into overlapping snippets.
This settings adjusts how much those snippets overlap (overlap tokens = snippet tokens / overlap ratio).
In general you shouldn't need to adjust this.

🔠 Language Model Settings

Avoid Repetition

Response Format

———

Max Output Tokens

The maximum number of tokens to generate in the completion. Increase to generate longer responses.

Creativity (aka Sampling Temperature)

Higher values allow the LLM to take more risks. Try values larger than 1 for more creative applications or 0 to ensure that LLM gives the same answer when given the same user input.

Answer Outputs

How many answers should the copilot generate? Additional answer outputs increase the cost of each run.

Run cost = 5 credits

Breakdown: 4 (Mixtral 8x7b Instruct v0.1 (Mistral)) + 3/run + 1 (lipsync)

By submitting, you agree to Gooey.AI's terms & privacy policy.

Show Raw Output

Assistant

Sure, based on the information provided, I understand that you are a community health worker at a patient's home and I am a nurse assisting you over the phone. Here's a summary of the encounter:

Symptoms:
You have not mentioned any specific symptoms the patient is experiencing. However, you should observe and take note of any symptoms the patient reports, such as fever, headache, vomiting, or abdominal pain.

Observations:
During the examination, you have reported the following observations:

Blood pressure: (waiting for measurement)
SPO2: (waiting for measurement)
Temperature: (waiting for measurement)
Urine sample: (waiting for results)

Recommendations:
Based on the information provided, I recommend the following:

Measure the patient's blood pressure, SPO2, and temperature.
Collect a urine sample and perform a malaria RDT if the patient is experiencing fever or other symptoms that may indicate malaria.
If the patient is experiencing any symptoms that may indicate pree

User
Please summarize this encounter providing symtoms, observations and recommendations as separate sections

💁‍♀️ Sources

Generated in 40.7s on

...

ℹ️ Details

How to Use This Recipe

🙋🏽‍♀️ Need more help? Join our Discord