Back to AI Providers

Soniox Setup Tutorial

Overview

This tutorial will guide you through setting up Soniox for use with Sokuji. Soniox is a real-time speech-to-speech translation service that transcribes and translates across 60+ languages in a single streaming API call.

1Create a Soniox Account

Go to console.soniox.com/signup and create a free account. If you already have a Soniox account, sign in directly at console.soniox.com.

2Open the Console

After signing in, open the Soniox Console. Your account starts with a default project ("My First Project").

3Go to API Keys

In the Console, open your project and click the API Keys section in the sidebar.

4Create an API Key

Click Create new key, give it a name (e.g. "Sokuji"), and confirm. Copy the key immediately — it is shown only once.

5Configure in Sokuji

Open Sokuji and select Soniox as your AI provider in the settings panel. Paste your API Key, choose the source language (or Auto Detect) and target language, pick a voice, and choose whether to output spoken translation or subtitles only.

6Test the Connection

Click the Start Session button and speak into your microphone to verify that real-time translation works. You should see the transcript and hear the translated audio (unless subtitles-only is enabled).

Features

What is Soniox?

Soniox is an end-to-end real-time speech translation service:

  • Real-time speech-to-speech translation: transcription, translation, and speech synthesis in one pipeline
  • Transcribe and translate 60+ languages in the same streaming API call, with translation at no extra cost
  • Automatic language detection: set the source to Auto Detect and Soniox identifies the spoken language
  • 12 multilingual voices for the translated speech output

Supported Languages

Supports 60+ languages with any-to-any translation between them:

  • The source language can be a specific language or Auto Detect
  • The target language can be any of the 60+ languages
  • Includes Chinese, English, Japanese, Korean, Spanish, French, German, Italian, Portuguese, Russian, and many more

Modes and Options

  • Spoken translation or subtitles only: toggle whether Soniox reads the translation aloud or shows text only
  • Voice selection: choose from 12 multilingual voices; one voice serves both directions
  • Two-way translation: translate both directions in a shared session in "Both" mode
  • Source language and translation mode: "Auto Detect" works when translating your own voice ("You" mode). For "Others" or "Both" mode, pick a specific source language — auto-detect cannot be reversed to translate the other side into your language.
  • No manual turn-detection setup — Soniox detects speech endpoints on the server

Troubleshooting

Common Issues

Authentication Error: Verify that your API Key is copied correctly with no extra spaces. If the key was lost, create a new one in the Soniox Console — keys are shown only once.

Connection Error: Check your network connection and ensure firewalls are not blocking WebSocket connections. If using a proxy, make sure it supports WebSocket.

No Translated Audio: If you only see subtitles and hear nothing, make sure spoken translation (not subtitles only) is enabled and that a voice is selected.

Tips for Best Results

  1. Use a stable network connection for the best real-time quality
  2. Use in a quiet environment for the best speech recognition
  3. Set the source language explicitly if auto-detection picks the wrong language
  4. Monitor your usage and balance in the Soniox Console

Need more help? Visit the Soniox documentation for details or check our GitHub repository for support.

Open Soniox Console