Interpretation built for real meetings
Many translation tools run a three-step pipeline: recognise, translate, synthesize. They end with a stock TTS voice. Linggua sends speech directly to a real-time interpretation model so the other side hears Chinese or English in your voice and at your pace.
They still hear you
Linggua clones the speaker’s voice by default. The other side hears a translation with your tone and pacing, not a generic machine voice.
- Clones the speaker’s voice by default
- No transcript-then-TTS stitching
Not tied to one meeting app
Install Linggua Microphone once, then select it as the microphone in your meeting app. Zoom, Teams, Tencent Meeting, Feishu, or the meeting app you already use. No plugin, no meeting bot and nothing for the other side to install.
- Virtual microphone, audio taken directly from the meeting app
- Other participants keep their usual meeting app; no meeting bot
- Use the meeting app’s audio as the source
Translate while you speak
Speech goes straight into the real-time interpretation model. It does not wait for a full transcript before starting. The translation moves with the back-and-forth pace of a real meeting.
- Real-time speech interpretation, not sentence-by-sentence waiting
- One less wait between recognition and synthesis
- Translates with sentence context
- Chinese and English, both directions
One audio path, three ways to listen
When they speak English, keep the original audio with Chinese captions, or listen to Chinese speech. When you speak Chinese, they hear English. Original audio, captions and translated speech can be combined instead of choosing only one.
- Original English + Chinese captions
- Chinese translated speech + bilingual captions
- Your Chinese speech → English for them
More than interpretation — a record of the meeting.
When the meeting ends, generate a bilingual summary covering discussion points, decisions and next steps. It is included in your subscription, with no extra charge and no translation minutes used.
- Bilingual summary, ready to forward to both sides
- Discussion points, decisions and next steps kept separate
- Included in the subscription, no translation minutes used
- Transcript used only to generate this summary, not kept long term
- Copy to clipboard
Names, terminology and fixed translations
Product names, customer names, industry terms and fixed translations can be configured. Interpretation and captions share the same vocabulary, so a term is not translated differently every time.
- Hot words are recognised as written
- Glossary entries fix the translation
- Interpretation and captions share one vocabulary
The difference from ASR → translate → TTS
| Linggua | Common stitched pipeline | Meeting bot | |
|---|---|---|---|
| Voice | Clones your voice by default | A fixed TTS voice | Usually a fixed voice or a host speaking |
| Path | Virtual microphone, audio from the meeting app, translating while you speak | Recognise, translate, synthesize in sequence | A bot joins the meeting and speaks |
| Meeting access | No specific meeting platform | Often a plugin or a specific platform | Depends on the platform allowing a bot |
| Other side | Installs nothing | May need a plugin or a specific platform | Usually no app, but they must allow the bot |
| Captions and speech | Original, captions and speech combine | Often text only or a fixed voice | Depends on the bot |
| Post-meeting summary | Built-in AI meeting summary | Usually absent, or raw transcript only | Depends on the bot |
Put your Chinese into an English meeting
Download Linggua for Mac and try 10 minutes free before choosing a plan.
Download for Mac