Pronunciation

Manage the global pronunciation dictionary of a speaker. Entries saved here apply to every future request for that speaker, unlike the per-request custom_pronunciations field on inference calls.

Per-request overrides

Include custom_pronunciations in any generate or stream request:

"custom_pronunciations": {
  "words": { "computer": "compyuter" },
  "phonemes": {}
}

Add word pronunciation

POSThttps://api.lucidweights.ai/v1/tts/pronunciation/{speaker}/words

Field	Type	Default	Description
word	string	required	The original word to replace
pronunciation	string	required	Replacement string
language	string	required	ISO 639-1 language code

Example

curl -X POST "https://api.lucidweights.ai/v1/tts/pronunciation/emily/words" \
  -H "x-api-key: lw_YOUR_SECRET" \
  -H "Content-Type: application/json" \
  -d '{"word":"computer","pronunciation":"compyuter","language":"en"}'

Add phoneme pronunciation

POSThttps://api.lucidweights.ai/v1/tts/pronunciation/{speaker}/phonemes

Field	Type	Default	Description
phoneme	string	required	Original phoneme to replace
replacement	string	required	Replacement phoneme
language	string	required	ISO 639-1 language code

Delete word pronunciation

DELETEhttps://api.lucidweights.ai/v1/tts/pronunciation/{speaker}/words/{word}

Query param language (default en).

Delete phoneme pronunciation

DELETEhttps://api.lucidweights.ai/v1/tts/pronunciation/{speaker}/phonemes/{phoneme}

Query param language (default en).