Unverified50% confidenceBenchmarkExact time
In a balanced ASR test covering 13 dialects, the Tunisian dialect achieved a word error rate (WER) of 0.478, the highest among all dialects
1
Sources
50%
Confidence
Long-term
Relevance
8/9/2026
First Seen
Sources
突尼斯方言NLP资源盘点：136个数据集的可用性真相
Related Claims
UnverifiedMultiple datasets labeled as 'Tunisian' actually originate from Moroccan dialect, or Tunisian constitutes only a very small portion of multi-dialect collections.65% similarUnverifiedIn general-purpose corpora such as Common Crawl, English typically accounts for more than 45%, while Portuguese accounts for less than 2%.60% similarUnverifiedFoneASR achieves a character accuracy rate of 88% in tests across 16 dialects, with first-character latency controlled at the hundred-millisecond level, and streaming recognition accuracy approaching offline recognition levels58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/715012API
curl https://kongchang.com/api/v1/knowledge/claims/715012MCP
get_claim(id=715012)