Get started
Translate live speech across supported languages with low-latency streaming.
Soniox lets you translate live speech in real time across 60+ languages and 3,600+ language pairs. You can stream translated text as people speak, or combine Soniox STT and TTS to build full speech-to-speech translation.
Speech translation is built into the Soniox real-time API, so you can start with transcription and enable translation with a simple configuration change.
Two ways to build speech translation
Translate live speech into written text.
Use the Soniox STT API to stream both the original transcript and translated text in real time. This is ideal for captions, subtitles, meeting translation, accessibility, agent assist, live dashboards, and multilingual transcription workflows.
Translate live speech into spoken output.
Combine Soniox STT and Soniox TTS to recognize speech, translate it in real time, and speak the result in the target language with low latency. This is ideal for live interpreters, bilingual voice agents, travel assistants, customer support, and real-time multilingual communication.
Two ways to translate live speech
Translate speech from any supported language into one target language.
Use one-way translation when many speakers or many languages need to be understood in a single output language, such as live captions, lectures, broadcasts, meetings, events, and customer calls.
Translate between two languages for live bilingual conversation.
Use two-way translation when both sides should speak naturally in their own language and understand each other in real time.
Language coverage
Speech translation uses the Soniox supported-language set and supports more than 3,600 language pairs.
See Supported languages for the language list and coverage details.
Run speech-to-text translation examples
Use the official examples repo to try real-time one-way and two-way translation with sample audio.
Get API key
Create a Soniox account and log in to the Console to get your API key.
API keys are created per project. In the Console, go to My First Project and click API Keys to generate one.
Export it as an environment variable:
Run speech-to-text translation examples
Choose your language and run the translation examples below.
Example | What it does | Output |
|---|---|---|
| Real-time one-way translation | Transcribes speech in any detected language and translates it into Spanish in real time. | Transcript + Spanish translation streamed together. |
| Real-time two-way translation | Transcribes and translates English ↔ Spanish in real time. Spanish → English, English → Spanish. | Transcript + bidirectional translations streamed together. |