> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://poly-ai.ferndocs.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://poly-ai.ferndocs.com/_mcp/server.

# Tutorial: Global ASR

> PolyAcademy Level 2 – Improve speech recognition accuracy with keyphrase boosting and transcript corrections.

> **Info**
>
> **Level 2 – Lesson 6 of 8** – Configure speech recognition to reduce misrecognition and improve routing.

> **Lesson**
>
> Level 2 · Intermediate · 20 min

[ASR](/agent-studio/voice/advanced-settings#keyphrases) determines what the agent *hears*. Every downstream system depends on it: retrieval, routing, response control, handoff logic. Small transcription errors here silently break otherwise well-designed agents. Global ASR is where you correct systematic listening problems at the source.

## What global ASR is for

#### Reduce clarifications

Minimize unnecessary clarification questions

#### Improve routing

Increase routing accuracy for calls

#### Stabilize retrieval

Improve Knowledge topic selection for spoken queries

#### Fix transcription

Correct recurring transcription errors at the source

Global ASR is not a substitute for good KB design or clear agent behavior rules. It improves transcription accuracy – it does not add new knowledge to the agent.

## Which tool to use

```mermaid
flowchart TD
    A[ASR problem identified] --> B{What type of error?}
    B -->|Word is correct but under-recognized| C[Keyphrase Boosting]
    B -->|ASR consistently hears the wrong thing| D[Transcript Corrections]
    C --> E[Add keyphrase → start with Default bias strength]
    D --> F[Add correction → define misheard expression and replacement]
    E --> G[Test in Call → increase bias only if errors persist]
    F --> G
    G --> H{Improvement confirmed?}
    H -->|Yes| I[Promote carefully]
    H -->|No| B
```

## Two essential tools

#### Keyphrase Boosting

Keyphrase Boosting biases the ASR model toward recognizing specific words or phrases more reliably.

### When to use it

#### Use cases

* A word is important for routing or actions
* Misrecognition causes the agent to take the wrong path
* Users consistently repeat themselves for the same term

#### Typical candidates

**Department names:**

* "front desk"
* "billing"
* "room reservations"

**Product or service names:**

* "bell desk"
* "late checkout"

**Action-critical verbs:**

* "cancel"
* "confirm"

### How to configure

#### Add the keyphrase

Enter it exactly as you want it recognized

#### Set bias strength

Start with Default bias strength

#### Adjust if needed

Increase toward Maximum only if errors persist

> **Warning**
>
> **Important guidance:**
>
> * Bias strength is cumulative. Over-biasing many terms can degrade overall accuracy
> * Avoid boosting generic words like "room" unless they are routing-critical
> * Prefer multi-word phrases over single words when possible
> * **Example:** If callers asking for the "bell desk" are frequently routed incorrectly, boost "bell desk" rather than boosting "desk"

#### Transcript Corrections

Transcript Corrections rewrite text after transcription, mapping common errors to the intended phrase.

### When to use it

#### Use cases

* ASR consistently hears the same wrong thing
* Accents or homophones cause predictable errors
* Brand or proper names are repeatedly mangled

#### Typical examples

**Homophones:**

* "copper" → "hopper"

**Accent-driven substitutions:**

* "reef fund" → "refund"

**Brand names:**

* "Hopper Consumer" → "Hopper"

### How to configure

#### Choose match type

Select substring or regex

#### Define misheard expression

Enter what ASR incorrectly transcribes

#### Define correction

Enter the intended replacement text

> **Warning**
>
> **Important guidance:**
>
> * Prefer narrow matches over broad regex
> * Avoid corrections that could unintentionally rewrite unrelated phrases
> * If unsure, test in Sandbox with real call audio before promoting
> * **Example:** If "cancel" is sometimes transcribed as "council," add a correction rather than increasing bias on "cancel" alone

#### Check your understanding

#### A caller says 'I need to cancel my booking' but the agent hears 'I need to council my booking' and routes to the wrong topic. Should you use Keyphrase Boosting or Transcript Corrections?

* Keyphrase Boosting – increase bias on 'cancel'
* Transcript Corrections – map 'council' to 'cancel'
* Both – boost 'cancel' and correct 'council' simultaneously
* Neither – this is a Managed Topic naming issue

**Answer:** Transcript Corrections – map 'council' to 'cancel'

ASR is consistently mishearing 'cancel' as 'council' – this is a predictable substitution error. Transcript Corrections rewrite after transcription and are the right tool for this pattern.

## A practical workflow

#### Review recent calls

Look for calls where:

* The agent asked for repetition
* The wrong KB topic was selected
* Routing or handoff failed unexpectedly

#### Identify the problem

Find the specific word or phrase that failed

#### Choose your tool

Decide:

* **Boost** if the word is correct but under-recognized
* **Correct** if the word is consistently misheard as something else

#### Apply minimal change

Make the smallest possible adjustment

#### Test in Call

Verify the improvement

#### Promote carefully

Only promote after confirming improvement

## Common mistakes

> **Warning**
>
> **Avoid these pitfalls:**
>
> * Boosting everything "just in case"
> * Using Transcript Corrections to fix KB design problems
> * Applying maximum bias globally without testing
> * Forgetting that local (flow-level) ASR settings override global ones

#### Check your understanding

#### When should you use Transcript Corrections instead of Keyphrase Boosting?

* When a word is under-recognized but correctly transcribed
* When ASR consistently mishears one specific thing as another
* When overall transcription quality is low across all words
* When you need multi-language support

**Answer:** When ASR consistently mishears one specific thing as another

Keyphrase Boosting improves recognition of words that are correct but under-weighted. Transcript Corrections rewrite after transcription – for consistent mishear patterns like "refund" being transcribed as "reef fund".

## Verification checklist

> **Tip**
>
> Verification is complete when:
>
> * Previously misheard phrases are transcribed correctly
> * The agent asks fewer clarification questions for known terms
> * Routing and KB selection stabilise in live call testing
> * Behavior remains consistent after promotion between environments
>
> Global ASR tuning is iterative. Small, deliberate adjustments here pay off more than large rewrites elsewhere.

## Try it yourself

#### Challenge: Diagnose and configure an ASR fix

A caller says "I'd like a refund" but ASR consistently transcribes it as "I'd like a reef fund". The `refund_request` topic therefore never triggers.

Answer:

1. Should you use Keyphrase Boosting or Transcript Corrections – and why?
2. Write the full configuration.

#### Hint

Ask yourself: is ASR hearing the right word but recognizing it poorly, or is it consistently mishearing one thing as another?

#### Example solution

1. **Transcript Corrections** – because ASR is consistently mishearing "refund" as "reef fund". This is a predictable substitution error, not an under-recognition problem.

2. **Configuration:**
   * **Match type:** Substring
   * **Misheard expression:** `reef fund`
   * **Correction:** `refund`

#### Check your understanding

#### What is the main risk of boosting too many keyphrases?

* Keyphrases stop working after 30 days
* Bias strength is cumulative and can degrade overall ASR accuracy
* It causes noticeable latency spikes
* Boosted words are excluded from Transcript Corrections

**Answer:** Bias strength is cumulative and can degrade overall ASR accuracy

Bias strength is cumulative – boosting too many terms can degrade overall ASR accuracy. Always start with Default bias and increase only if errors persist.

#### [← Previous: Audio management](/academy/poly-academy/poly-academy-level-2/audio-management)

Lesson 5 of 8

#### [Next: Variants →](/academy/poly-academy/poly-academy-level-2/variants)

Lesson 7 of 8