Fix pronunciations

TTS models guess at names, brands, and jargon — and one wrong name ruins the take. A workspace dictionary entry fixes the word once; the Normalizer applies it on every run from then on. Needs the dictionary:read / dictionary:write scopes.

Add an entry

Two methods:

• spelled — a phonetic respelling (“Podonos” → “poh-DOH-nohs”)
• recorded — reference audio of the word, from an upload

1
client.dictionary.create_dictionary_entry(
2
    word="Podonos",
3
    method="spelled",
4
    pronunciation="poh-DOH-nohs",
5
    language="en-us",
6
)

Not sure how to respell it? Ask for a suggestion first:

1
s = client.dictionary.suggest_pronunciation(word="Podonos", language="en-us").data

Entries are per-language — a name said one way in English and another in Korean gets one entry per language.

Review what’s in the dictionary

1
for e in client.dictionary.list_dictionary_entries(language="en-us").data:
2
    print(e.word, "→", e.pronunciation)
3
4
client.dictionary.search_dictionary_entries(search="podo")
5
client.dictionary.list_dictionary_languages()    # languages you can add entries for

Dictionary management is SDK/API only for now — the CLI doesn’t cover it yet.

Where it takes effect

The Normalizer applies dictionary entries while rewriting text into spoken form — so the fix reaches every workflow in the workspace that has a Normalizer node, current and future. If a line still comes out wrong, the Accuracy validator is the safety net: it hears the mispronunciation as a word miss and regenerates the line.

Related

• Normalizer — where entries are applied
• Validate every line — catching what slips through
• Upload a file — reference audio for recorded entries