Published • loading... • Updated
Humyn Labs launches report for AI voice benchmarking across Global South languages
The benchmark tests 15 commercial models across 22 languages and says global rankings do not reliably predict regional speech-recognition performance.
- Humyn Labs released the BRIDGE benchmark to evaluate 15 commercial AI speech-recognition models across 22 non-English languages spoken by over 5.5 billion people in the Global South using real conversational data.
- ASR providers previously published accuracy scores using English-first, internet-trained datasets with little independent validation, misleading enterprises making million-dollar deployment decisions, said Manish Agarwal, co-founder, Humyn Labs.
- Built on real two-person conversations rather than scripted audio, the benchmark applies a seven-metric scoring stack including Semantic Similarity and Phoneme-Informed Error Rate beyond standard word error measures.
- ElevenLabs Scribe v2 leads globally at 10.6% word error rate, yet rankings obscure regional performance; AssemblyAI Universal ranks third globally but outperforms all competitors in Vietnamese.
- Single aggregate benchmark scores cannot reliably guide cross-regional deployment decisions, Ishank Gupta, co-founder, noted; the full dataset is available on Hugging Face for independent analysis.
Insights by Ground AI
54 Articles
54 Articles
Coverage Details
Total News Sources54
Leaning Left1Leaning Right0Center41Last UpdatedBias Distribution98% Center
Bias Distribution
- 98% of the sources are Center
98% Center
C 98%
Factuality
To view factuality data please Upgrade to Premium
