---
format: "aidr-story-markdown/v1"
id: "8f8c726fbdb30a61c539a826b3f2c20aa1af7074558cd7c31eba67c89b69ee35"
canonical_url: "https://aidr.today/8f8c726f?lang=en"
title: "Code-Switching Detection via Language-Selective Dimensions in Speech Foundation Models"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-14T00:00:00.000Z"
category: "Research"
topics: ["openai","inference"]
source_urls: []
summary: "A lightweight method detects code-switching directly from speech using frame-level language information in pretrained speech encoders, achieving a mean F1 of 0.798 across 14 language pairs in CS-FLEURS with the Whisper-small encoder."
---

# Code\-Switching Detection via Language\-Selective Dimensions in Speech Foundation Models

> [Open the canonical story](<https://aidr.today/8f8c726f?lang=en>)

**Published:** 2026-10-14T00:00:00.000Z
**Category:** Research
**Topics:** openai, inference

## Summary

A lightweight method detects code\-switching directly from speech using frame\-level language information in pretrained speech encoders, achieving a mean F1 of 0\.798 across 14 language pairs in CS\-FLEURS with the Whisper\-small encoder\.

## Sources

_No safe source URL was recorded._