The Scenario
You are building a voice assistant for a medical client. The assistant needs to correctly pronounce 40 specialized terms — drug names, anatomical terms, procedure names — that the default voice model mangles. You have an Excel workbook with two columns: the term and its phonetic pronunciation. The client gave it to you last week.
To get these into Vapi, you need to create a pronunciation dictionary resource via the 11Labs provider integration. Creating it from a 40-row worksheet by hand or by script is not how you planned to spend your Thursday afternoon.
The bad version:
- Read the Vapi API docs for 11Labs provider resources, find the pronunciation dictionary endpoint
- Construct the request body manually — figure out the exact JSON structure for 40 entries, copy each term and phonetic from your workbook into the JSON
- Authenticate and fire the API call, parse the response to confirm it worked
- Realize on your second attempt that entry 23 had a formatting error in the phonetic field and the whole dictionary needs to be recreated
The client demo is next week. You have other things that need your attention today.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that lives inside your Excel workbook. It reads your term and phonetic columns directly from the worksheet and creates the Vapi 11Labs pronunciation dictionary resource through the built-in integration.
Take the 40 medical terms and phonetics in my Excel sheet and create a single Vapi pronunciation dictionary resource with all of them
What You Get
- A single Vapi 11Labs pronunciation dictionary resource created with all 40 term-phonetic pairs
- SheetXAI reads the entries directly from your workbook — no JSON construction, no manual API call
- The dictionary ID comes back in the response so you can assign it to the assistant immediately
What If the Data Is Not Quite Ready
Some rows in the workbook are header rows or blank rows that should be skipped
Read my Excel sheet where column A has the term and column B has the phonetic pronunciation — skip the header row and any row where column A is blank — create a Vapi 11Labs pronunciation dictionary resource from all remaining entries
Some phonetic strings in column B have extra quotation marks from a paste
Read my Excel sheet where column A has the term and column B has the phonetic pronunciation — strip any extra quotation marks from column B values — then create a Vapi 11Labs pronunciation dictionary resource with all entries
You need to split the 40 terms into two dictionaries by category, using column C as the category label
Read my Excel sheet where column A is the term, column B is the phonetic pronunciation, and column C is the category — create one Vapi 11Labs pronunciation dictionary named 'Drug Names' for all rows where column C is 'drug', and a second dictionary named 'Anatomical Terms' for all rows where column C is 'anatomy'
Validate, create, and confirm in one shot
Read column A for terms and column B for phonetic pronunciations in my Excel sheet — skip any row where either column is empty — create a Vapi 11Labs pronunciation dictionary resource with all valid entries — then write the returned dictionary ID into cell D1 and the total entry count into cell D2
Combining the validation, the creation, and the confirmation write-back in one prompt means you finish with a record of exactly what was created, ready to hand to the client.
Try It
Get the 7-day free trial of SheetXAI and open the Excel workbook where your pronunciation terms live, then ask it to create the Vapi dictionary resource from your rows. For related tasks, see how to bulk update assistant configs or export your assistant inventory.