Peer-reviewed research
HiACC: Hinglish adult & children code-switched corpus
Shruti Singh et al., Data in Brief (via PubMed Central), 2025
Read the originalRead on October 17, 2026
- What it is
- A 2025 data paper that introduces a Hinglish (Hindi-English) speech corpus for adults and children. Its introduction cites recent evaluations of how much word error rate rises for code-switched speech.
- How much weight it carries
- Peer-reviewed data paper. The 30 to 50 percent range is a figure the authors cite from other evaluations and do not reproduce here, so treat it as a literature range, not a measurement of any one engine.
Facts taken from this source
- Recent evaluations, which the HiACC authors cite, show that ASR models have a 30 to 50 percent relative increase in word error rate on code-switched speech compared with monolingual input.
“Recent evaluations have shown that ASR models experience a relative increase in Word Error Rate (WER) of 30–50 % when exposed to code-switched speech compared to monolingual input” (Abstract)