Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
80 changes: 80 additions & 0 deletions fern/customization/deepgram-keyword-boost.mdx
Original file line number Diff line number Diff line change
@@ -0,0 +1,80 @@
# Deepgram Keyword Boosting

Deepgram’s keyword boosting feature allows further customization to improve the transcription accuracy. This is particularly useful in customer service scenarios where specialized terminology, product names, or customer names are frequently misrecognized.

## Why Use Keyword Boosting?
Keyword boosting is beneficial for:

- Enhancing the recognition of customer names and brand-specific terms.
- Improving transcription accuracy in outbound sales, inbound support, and ordering systems.
- Reducing errors in transcriptions of product or service names.

## API Call Integration
To enable keyword boosting, add a `keywords` parameter to your Vapi assistant’s `transcriber` section. This should include the keywords and their respective intensifiers.

### Example of POST Request
```bash
curl \
--request POST \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '{
"name": "PizzaBot",
"model": {
"model": "gpt-4o",
"provider": "openai"
},
"voice": {
"voiceId": "pizza-assistant",
"provider": "azure"
},
"transcriber": {
"provider": "deepgram",
"model": "nova-2",
"language": "en",
"smartFormat": true,
"keywords": [
"Margherita:3",
"Pepperoni Deluxe:3",
"BBQ Chicken:3"
]
},
"firstMessage": "Welcome to Pizza Express! What would you like to order?",
"firstMessageMode": "assistant-speaks-first"
}' \
https://api.vapi.ai/assistant
```

### Explanation of Parameters:
- `name`: The name of the assistant.
- `model`: Specifies the model and provider for the assistant’s conversational capabilities.
- `voice`: Specifies the voice and provider for the assistant’s speech.
- `transcriber`: Uses Deepgram for transcription, with custom keywords for pizza menu items.
- `firstMessage`: The initial message the assistant will speak.
- `firstMessageMode`: Specifies that the assistant speaks first.

## Intensifiers
Intensifiers adjust how strongly the model prioritizes certain words:

- **Boosting Example:** `keywords=Margherita:4` (higher likelihood of recognition)
- **Suppressing Example:** `keywords=Anchovies:-2` (lower likelihood of recognition)

## Common Use Cases
### 1. Pizza Order System
Imagine you’re building a voice agent for a pizza delivery service. Customers frequently order pizzas with specific names that may not be easily recognized by default speech models. You can use keyword boosting to ensure accurate transcription of menu items like "Margherita," "Pepperoni Deluxe," or "BBQ Chicken."

### 2. Financial Services
In banking and finance, customers may mention account types, transaction codes, or financial products. Using keyword boosting ensures that terms like "ACH Transfer," "FICO Score," or "401k" are transcribed accurately.

### 3. Healthcare Support
Medical professionals and patients might use technical medical terms during calls. Boosting keywords such as "Amoxicillin," "MRI Scan," or "Hypertension" helps maintain accuracy in transcription.

### 4. Travel and Hospitality
When booking flights or hotels, recognizing destination names and travel jargon correctly is crucial. Boosting city names like "Honolulu," "Santorini," or "Maldives" ensures better accuracy.

## Best Practices for Keyword Boosting
- **Use Customer-Focused Keywords:** Common product names, store locations, and customer names.
- **Set Balanced Intensifiers:** Avoid extreme values that could cause false positives.
- **Limit Excessive Keywords:** Overloading with too many keywords may reduce effectiveness.

By implementing keyword boosting, customer service applications can significantly improve transcription accuracy, leading to better user experiences.
21 changes: 13 additions & 8 deletions fern/docs.yml
Original file line number Diff line number Diff line change
Expand Up @@ -86,6 +86,9 @@ analytics:
navigation:
- tab: documentation
layout:
- link: Status Page
href: https://status.vapi.ai/

- section: Getting Started
collapsed: true
contents:
Expand Down Expand Up @@ -119,6 +122,9 @@ navigation:
- page: Python Outbound Snippet
path: examples/outbound-call-python.mdx

- page: FAQ
path: faq.mdx

- section: Build
contents:
- section: Assistants
Expand Down Expand Up @@ -169,8 +175,13 @@ navigation:
- page: Tavus
path: customization/custom-voices/tavus.mdx

- page: Custom Keywords
path: customization/custom-keywords.mdx
- section: Custom Keywords
contents:
- page: Introduction
path: customization/custom-keywords.mdx
- page: Deepgram Keyword Boosting
path: customization/deepgram-keyword-boost.mdx

- page: Custom Transcriber
path: customization/custom-transcriber.mdx

Expand Down Expand Up @@ -448,12 +459,6 @@ navigation:
- page: Glossary
path: glossary.mdx

- page: FAQ
path: faq.mdx

- link: Status Page
href: https://status.vapi.ai/

- section: Legal
collapsed: true
contents:
Expand Down