;
};
**Level 2 – Lesson 5 of 8** – Understand and manage the audio cache for optimal performance.
[Audio Management](/voice-channel/audio-library) controls how TTS output is generated, cached, and replayed. Without understanding caching, teams often think changes "didn't apply" when they're hearing old audio.
## Understanding the audio cache
Cached audio stores previously generated TTS so it can be replayed instantly, reducing latency and keeping repeated phrases consistent.
Audio is only cached if the **same utterance is generated at least twice in a 24-hour window**.
One-off utterances will *not* persist in cache by default.
## Managing cached audio
Navigate to **Voice > Audio library** in the platform.
Check the list of cached utterances:
* Greeting
* Transfer / handoff language
* SMS offer phrasing
* Closings and confirmations
For any high-frequency utterance:
* Open it and review how often it has been used
* Adjust **stability** and **clarity** *for that utterance only* if needed
* Use the **play** button to preview changes
If an utterance must remain stable:
* Generate it multiple times in 24 hours, or
* Upload a static audio file to overwrite the cached version
## Check your understanding
## Interaction style (response latency)
Interaction style controls how quickly the agent responds after detecting user speech. This directly affects interruption rate and perceived naturalness.
**\~400ms latency**
Extremely fast, higher interruption risk.
**\~1200ms latency**
Prioritizes speed.
**\~1600ms latency**
Default for most use cases.
**\~2000ms latency**
Slower, more deliberate, fewer interruptions.
## Barge-in
Barge-in determines whether callers can interrupt the agent mid-speech.
* Useful for Turbo mode
* Can feel chaotic if enabled without careful phrasing and latency tuning
## Pronunciations
Ensure domain-specific terms are spoken clearly and correctly in Call.
Pronunciations are defined in the **Pronunciations** tab under **Voice > [Response Control](/voice-channel/advanced/pronunciations)** and applied globally. They modify how text is converted to speech, without changing the underlying text.
### When to use pronunciations
Product names that are mispronounced
Locations, people, departments
Structured read-back requirements
Phrases where pacing matters for comprehension
### How pronunciations work
Matching is done using **regular expressions**. Replacements can be:
**International Phonetic Alphabet**
For precise pronunciation control
**Speech Synthesis Markup Language**
Such as `` for pauses
**Pattern matching**
`\1`, `\2`, etc. for reformatting
### Examples
**Regex:** `\bLouvre\b`
**Replacement:** `/ˈluːvrə/`
**Case sensitive:** `FALSE`
**Regex:** `(\d{3})[ -]?(\d{3})[ -]?(\d{4})`
**Replacement:** `\1 \2 \3`
### Best practices
Add pronunciations one at a time
Test each change in Call before adding more
Prefer clarity over complexity – overly complex regex is hard to maintain
## Check your understanding
## Verification checklist
**After any voice or phrasing change:**
* Start a *new* call session
* Confirm you are hearing updated audio, not a cached version
* Validate that turn-taking still feels natural after changing latency or barge-in
* Mispronounced terms are corrected consistently
* Pauses improve comprehension rather than slowing the call excessively
## Try it yourself
Your agent says "Hopper" but it is consistently pronounced incorrectly (sounds like "Hooper"). You also want phone numbers read back with a natural pause between each segment.
Write both pronunciation configurations:
1. IPA correction for "Hopper"
2. Phone number formatting with 0.5s pauses
For the IPA, write out what "Hopper" sounds like phonetically. For the phone number, use regex capture groups to split the digits and insert SSML `` tags.
**Brand name correction:**
* **Regex:** `\bHopper\b`
* **Replacement:** `/ˈhɒpər/`
* **Case sensitive:** FALSE
**Phone number with pauses:**
* **Regex:** `(\d{3})[ -]?(\d{3})[ -]?(\d{4})`
* **Replacement:** `\1 \2 \3`
## Check your understanding
Lesson 4 of 8
Lesson 6 of 8