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

# AI Speaker pronunciation tips

> Fix mispronounced names, acronyms, and homographs in AI Speaker audio by adjusting spelling and syllable emphasis.

<Note>
  While AI Speakers perform well in most cases, achieving perfect pronunciation can require some finesse and may involve multiple attempts. Areas where you might need to make adjustments include:
</Note>

* Names with unique spellings or pronunciations.
* Pronouncing individual letters for abbreviations or initialisms.
* Adjusting syllable emphasis for different forms of a word (e.g., noun vs. verb).
* Modifying the pronunciation of homographs—words that are spelled the same but pronounced differently.

To refine pronunciation:

1. Modify the word’s spelling in the script. Experiment with phonetic spellings until you achieve the desired pronunciation.
2. [Convert the Speaker clips to audio](/ai-speech/covert-to-audio) from the [Timeline](/timeline/timeline-overview).
3. Highlight the section on converted speaker audio in your script.
4. [Use the correction tool](/script-editing/correct-your-transcript) to adjust the transcript spelling without affecting the script media.

## Timing and intonation of your AI Speaker audio

There are a few things to keep in mind that will affect the timing and intonation of generated AI Speaker audio:

* Punctuation affects the spacing of your generated AI Speaker audio.
* Generating more than one word can improve the natural sound of your AI voice.
* You can adjust the [clip speed](/visuals/clip-properties) and volume to improve the natural sound of your AI Speaker audio.


## Related topics

- [Troubleshooting and using AI Speakers](/ai-speech/troubleshooting.md)
- [Generate text-to-speech audio](/ai-speech/tts.md)
- [Create a custom voice clone](/ai-speech/custom-speaker.md)
- [Create a first-draft video with Quick Design](/ai-assistant/quick-design.md)
- [Speakers](/script-editing/speakers.md)
