fix: no auto detection, precision of 1 language and only

Author: egenthon-cmd

chapters/pre-recorded-stt/quickstart.mdx

@@ -109,6 +109,10 @@ transcription = gladia_client.transcribe(
 ```
 </CodeGroup>
 
+<Note>
+  With `"solaria-3"`, set **one language** in `language_config.languages` — for example `["fr"]`. Do not pass multiple languages or enable code switching.
+</Note>
+
 <Note>
 Want to go further? See [Audio Intelligence](/chapters/pre-recorded-stt/audio-intelligence) for add-ons like:
 - [Speaker diarization](/chapters/audio-intelligence/speaker-diarization): separate the speakers across the conversation
@@ -277,7 +281,7 @@ const response = await fetch("https://api.gladia.io/v2/pre-recorded", {
     audio_url: "YOUR_AUDIO_URL",
     model: "solaria-3",
     language_config: {
-      languages: [],
+      languages: ["fr"],
     },
     diarization: true,
     diarization_config: {
@@ -318,7 +322,7 @@ curl --request POST \
   "audio_url": "YOUR_AUDIO_URL",
   "model": "solaria-3",
   "language_config": {
-    "languages": []
+    "languages": ["fr"]
   },
   "diarization": true,
   "diarization_config": {

snippets/choose-model.mdx

@@ -1,7 +1,7 @@
 <Tip>
 Pass `model` in your pre-recorded request body to choose the transcription model:
 
-- `"solaria-3"` — our latest model (best-in-class on EU languages). **Async (pre-recorded) only** — not available for live transcription. **No code switching.**
+- `"solaria-3"` — our latest model (best-in-class on EU languages). **Async (pre-recorded) only** — not available for live transcription. **Single language only** — pass exactly one language in `language_config.languages` (no code switching or multi-language configuration).
   - **Languages:** English, French, Spanish, German, Dutch, Italian, Portuguese
   - **Add-ons:** summarization, translation, PII redaction, and audio to LLM
 - `"solaria-1"` — our generalist model (code switching, 100 languages). Available for async and live.