Analysis
Microsoft AI released MAI-Transcribe-2 this week, pricing the speech-recognition model at $0.10 per audio hour through the end of 2026 -- a 72% cut from the $0.36-per-hour launch price of its predecessor five months ago, according to Microsoft's own announcement and VentureBeat. Microsoft claims the model is up to 10 times faster than OpenAI's GPT-Transcribe, 7 times faster than ElevenLabs' Scribe v2, and 5 times faster than Google's Gemini 3 on transcription tasks, while expanding language coverage to 60 and adding speaker diarization and word-level timestamps.
The model is available in public preview through Azure Speech, Microsoft Foundry, the MAI Playground, and third-party router OpenRouter -- distribution that puts it in front of both enterprise Azure customers and independent developers comparing models head-to-head. For a business processing 100,000 hours of call-center audio annually, Microsoft says the price cut takes the bill from $36,000 to $10,000, a saving concrete enough to force competitive responses rather than just marketing claims.
Racing to commoditize transcription before rivals catch up
Speech-to-text has quietly turned into one of the more commoditized corners of the AI stack, with OpenAI, Google, ElevenLabs, Deepgram, and AssemblyAI all competing on largely similar accuracy bands -- which is exactly why price and speed, not raw capability, are now the primary battlegrounds. Microsoft cutting its own price 72% inside five months suggests the company is willing to run transcription near cost to keep developers inside the Azure ecosystem, mirroring how AWS and Google Cloud have historically used loss-leading infrastructure pricing to win platform lock-in rather than direct product profit.
The performance claims -- 10x, 7x, 5x faster than named competitors -- come from Microsoft's own benchmarking, not an independent third party, a caveat that applies to essentially every speed comparison vendors publish in this market. ElevenLabs and OpenAI have not yet responded with matching price cuts, and whether they follow Microsoft down in price or instead compete on accuracy and language-specific quality will determine whether this becomes a margin-destroying price war across the entire category or stays a Microsoft-specific loss leader tied to Azure lock-in.
