Translate a document
Create a new language version. Credits match MP3 export.
Translate in the More menu creates a new document version in another language. It bills the same credits as MP3 export and it is separate from New version → Original (that path still uses the language model).
Create a translated version
- Open the document in the reader.
- Open More in the side toolbar.
- Click Translate.
- Pick a target language from the list.
- Confirm. Speechdash adds a completed version named after that language.
The list is the overlap between the translation service and the languages Speechdash already supports for playback. The first time translation starts, language models may download before the list appears. That can take several minutes.
Timestamps
Transcript clocks are playback references, not text to translate. Speechdash copies them into the new version unchanged:
- Time-only headers such as
**0:34** - Speaker headers such as
**SPEAKER_00** · 0:34 - Inline cues such as
(0:34)
The translation engine never sees those tokens, so it cannot reformat or duplicate them. The in-app Show timestamps toggle still hides them in the reader only.
Credits
Translation uses the same rate as MP3 export: 0.5 credits per started 30 seconds of estimated speech (minimum 0.5). The dialog shows the estimate with the coins icon. Listening to the new version with Cloud voice still uses the usual speech credits.
If translation fails after the charge, Speechdash refunds the credits.