This article explains how to change how AI reads text aloud during audio playback. There are two methods.
Corrections you make are unified across the entire team and will be reflected in the reading of other SOPs as well.
① Correct misreading while viewing an SOP
- Click the "Correct misreading" icon
- Select the relevant text (in this example, MΩ)
-
Enter the correct reading and click "Register". When you open it again, the specified reading will be displayed.
※By default, users at Group Administrator level and below will show a "Report" status, and the correction will be reflected once a Team Administrator or higher approves it.
② Set up in advance from Dictionary / Templates
You can also configure this by opening "Dictionary / Templates" from the sidebar.
③ Bulk import with Excel or CSV
When you want to register or update a large number of readings, you can import them using an Excel or CSV file from the "Bulk import" button in the "Pronunciation Dictionary" tab. This is convenient when you want to apply a glossary or in-house reading list all at once.
- Open "Dictionary / Templates" from the sidebar and switch to the "Pronunciation Dictionary" tab
- Click "Bulk import" in the upper right
- Click the "Excel template" button in the upper right to download the template. If readings are already registered, they will be included in the download, so you can edit and re-upload them as is
- Drag and drop the template (or your own CSV / Excel with the columns "Notation / Reading / Language (Optional)"), or paste the contents
- A Preview of changes is displayed, and you can confirm the contents in Add / Update / Skip / Error categories
- If everything looks good, click "[Register N items]" to confirm
- Supported file formats: CSV, TSV, Excel (.xlsx / .xlsm). Character encoding is automatically detected as UTF-8 or Shift-JIS
- Language column: If omitted, it will be registered as Japanese (ja)
- Japanese readings must be entered in hiragana or katakana. Any other characters (kanji, letters, symbols only, etc.) will result in an error
- Readings that are the same as existing content will be automatically skipped
- After registration, it may take a few minutes for the changes to be reflected in existing audio (regenerated)
- If a user at Group Administrator level or below performs bulk import, a Report will be sent to Team Administrators or higher to request approval, just like a regular addition
Related dictionary features
"Dictionary / Templates" also contains the following dictionaries. All of them support bulk import using Excel or CSV.
- Notation Unification Dictionary: Text replacement rules within your team (Reference: Unify notation variations (Notation Unification Dictionary))
- Translation glossary: Pre-specify translations for auto-translation (Reference: Translation glossary)