Captions show text; speech translation plays audio
Transcription turns speech into text in the same language. Translated captions show that content in another language. Speech translation plays the translation aloud. Keep the original audio with translated captions to hear the speaker’s voice, or choose translated speech to listen in your language.
Three listening modes in Linggua
Choose a mode under “Their voice”. The current product focuses on Chinese–English translation; these examples assume the other person is speaking English.
| Mode | What you hear | What you read |
|---|---|---|
| Original voice + captions | Original English audio | English source text and Chinese translation |
| Translated voice + captions | Chinese speech; original audio is muted | Source text and translated captions |
| Original voice only | Original English audio | No new incoming captions |
Original audio can still have translated captions
Original voice + captions preserves the speaker’s pronunciation, tone and pace while displaying a Chinese translation. The meeting app plays the original audio. Generating captions and playing that original audio are separate processes.
Translated speech replaces the original audio
Translated voice + captions plays Chinese speech and mutes the incoming original audio to avoid playing both languages together. Stopping the translation connection restores the original meeting audio. Changing listening modes reconnects the translation service.
Listening and speaking are independent
The listening mode controls what you hear. Your microphone mode controls what the other participants hear: Translate sends translated speech, Original voice sends your own voice, and Muted stops sending live microphone audio.
Choose a mode for the conversation
Use original audio with captions when pronunciation, tone or the exact wording matters. Choose translated speech when you prefer to listen to the translation. Use original audio only when you do not need incoming translation. For names, dates and numbers, refer to the source captions and confirm details with the speaker.