Languages and localization
Document language, listening language, question language, and supported locales.
Speechdash supports many languages for import, classification, playback, and version generation. Several language settings work together; they are not always the same value.
Audio and video transcription
Drag & Drop
MP3, MP4, M4A, MOV, AAC, WAV, OGG, OPUS, MPEG, WMA, WMV, PDF
or
Browse filesThree language concepts
| Concept | What it controls |
|---|---|
| Document language | How text is segmented, classified, and stored |
| Listening language | Which locale TTS uses when reading aloud |
| Question / STT language | Speech-to-text input and assistant reply language in Ask questions |
Document language
Set automatically when you import a PDF or text file. Speechdash uses script detection, lexicon hints, and language classification.
Change it later in Rename (side toolbar → More → Rename). Updating document language reindexes sentences and may invalidate cached audio.
Listening language
The voice popover on the playback bar lets you pick a listening language independent of the document language. Useful when:
- You read content in one language but prefer narration in another
- You study with bilingual material
- You generate an Original version translated to a target language
See the full list in Supported languages.
Chat and voice input languages
Under Settings → General:
- Voice input: defaults to document language when unset
- Assistant replies: defaults to match your question when unset
These affect Ask questions only, not playback.
Version generation languages
When creating a New version, pick an output language. Original can translate source text while keeping structure; AI versions (Natural, Lecture, Conversational) rewrite in the chosen language.
Currency and locale (account)
Account country and currency (EUR or USD) are set during onboarding or in Settings → Account. They affect billing display, not document languages.