> ## Documentation Index
> Fetch the complete documentation index at: https://docs.rime.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Pronunciation control

> Check dictionary coverage, specify custom pronunciations, and request dictionary additions to control exactly how Rime speaks your words.

Rime's models power the voice of your product, and pronunciation control makes sure brand names, product names, and domain-specific words sound right before they reach your users.

There are three ways to control pronunciation:

1. **Check dictionary coverage** to see whether a word is already pronounced correctly out of the box.
2. **Specify a custom pronunciation inline** using the Rime phonetic alphabet.
3. **Request a dictionary addition** from the Rime team.

***

## Check dictionary coverage

Use the [Coverage API](/api-reference/other/oov) to determine which words in your text are not yet in Rime's pronunciation dictionary. Words that aren't covered will still be spoken — the model predicts a pronunciation, which is often suitable when the spelling follows typical linguistic patterns — but it's worth verifying uncommon words before they reach end users.

## Specify a custom pronunciation inline

For Mist-family models, you can specify a custom pronunciation directly in your TTS request by passing a phonetic string in curly brackets and setting `phonemizeBetweenBrackets: true`. The strings use the [Rime phonetic alphabet](/platform/rime-phonetic-alphabet), which is inspired by IPA.

See [Custom pronunciation](/docs/custom-pronunciation) for the full reference.

### Generate the phonetic string from audio

You can write phonetic strings by hand from the alphabet reference, or let Rime generate one for you: record the word (or synthesize it) and post the raw audio to the [Phonemize API](/api-reference/other/phonemize).

```bash theme={null}
curl -X POST https://optimize.rime.ai/phonemize \
     -H "Authorization: Bearer $RIME_API_KEY" \
     -H "Content-Type: audio/wav" \
     --data-binary @speech.wav
```

The response contains the phonetic transcription of the audio:

```json theme={null}
{
  "audioId": "9b2d8ad2-0618-4e96-b255-98b2f7488061",
  "phonemeString": "h0El1o !",
  "authed": 1
}
```

Drop any trailing punctuation token and paste the string into your TTS request inside curly brackets — here, `{h0El1o}` — with `phonemizeBetweenBrackets: true`.

<Warning>
  [`phonemizeBetweenBrackets`](/docs/custom-pronunciation) works on **Mist v1 and v2** only. It is not yet supported on Mist v3, Coda, or Arcana. For brand or product name pronunciations on those models, request a dictionary addition, or respell phonetically in plain English (accepting that this is approximate).
</Warning>

## Request a dictionary addition

If a word is mispronounced and you'd like it fixed at the dictionary level, reach out to your account manager via Slack or email, or contact [sales@rime.ai](mailto:sales@rime.ai). Our linguists typically review and add new words within one week.

***

## Tips

* Check coverage regularly when working with new or domain-specific content.
* Brand names, technical terms, and made-up words are the most common sources of mispronunciation — verify these first.
* For flows where pronunciation control matters most, use Mist v1/v2 so you can correct pronunciations inline.