Microsoft releases MAI-Transcribe-1.5 with 2.4% word-error rate across 43 languages
Microsoft AI Introduces MAI-Transcribe-1.5: 2.4% WER on Artificial Analysis, Best-in-Class FLEURS Accuracy, and Up to 5x Faster Long-Audio Transcription
MarkTechPost
Summary
Microsoft AI released MAI-Transcribe-1.5, its second-generation speech-to-text model covering 43 languages with keyword biasing for domain-specific terms. It achieves 2.4% word-error rate on the Artificial Analysis leaderboard and transcribes an hour of audio in under 15 seconds. It is available in Azure AI Foundry.