PulseAugur
EN
LIVE 21:02:46

Microsoft AI launches MAI-Transcribe-1.5 with 43 languages, faster speeds

Microsoft AI has released MAI-Transcribe-1.5, an updated speech-to-text model supporting 43 languages and diverse acoustic conditions. The model achieves a 2.4% word error rate on the Artificial Analysis benchmark and boasts best-in-class accuracy on FLEURS. It also offers significant speed improvements, transcribing an hour of audio in under 15 seconds, and includes a keyword biasing feature to improve accuracy for domain-specific terms. AI

IMPACT Enhances transcription accuracy and speed for enterprise applications, potentially improving efficiency in content creation and analysis.

RANK_REASON This is a product update for an existing speech-to-text model, not a novel frontier model release.

Read on MarkTechPost →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

COVERAGE [2]

  1. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    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

    <p>Microsoft AI has released MAI-Transcribe-1.5, the second iteration of its in-house speech-to-text family. The model covers 43 languages, adds keyword (entity) biasing for domain-specific terms, posts a 2.4% Word-Error-Rate on the Artificial Analysis leaderboard, and transcribe…

  2. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    Microsoft AI has unveiled MAI-Transcribe-1.5, a speech recognition model achieving 2.4% word error rate across 43 languages. The model runs up to 5x faster than

    Microsoft AI has unveiled MAI-Transcribe-1.5, a speech recognition model achieving 2.4% word error rate across 43 languages. The model runs up to 5x faster than comparable models and can transcribe an hour of audio in under 15 seconds. Integrated into Copilot, Teams and GitHub. h…