Chinese-English Code-Mixed Speech with Prosodic Annotations

A 3-hour collection of naturalistic Chinese-English code-mixed speech sourced from social media videos. The dataset includes dual-level annotations, featuring manual token-level labeling for prosodic analysis at switch boundaries. It was created by hafsamenaz1 and last updated on Hugging Face in April 2026.

Use Cases

• Analyzing prosodic reset patterns at language switch boundaries based on the acoustic alignments mentioned in the description.
• Training models for automatic detection of code-switching points in speech based on prosodic features.
• Studying phonological shifts in naturalistic bilingual conversation based on the curated social media content.

Strengths

• 3 hours of curated, naturalistic speech data.
• Includes manual token-level annotations for precise acoustic alignment.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file formats are unknown, which may limit suitability assessment.

Provenance

Source

hafsamenaz1 on Hugging Face.

Collection Method

Sourced from diverse social media video content and manually annotated.

Freshness

Last updated 2026-04-25 07:51:07; freshness should be verified.