Microsoft releases real-time transcription model MAI-Transcribe-2-Streaming
microsoft ai released mai-transcribe-2-streaming, a real-time transcription model that transcribes 60 languages with high accuracy, allowing voice agents to respond quickly. this affects voice agent developers and companies using ai for customer service.
microsoft ai released two new models: mai-transcribe-2-streaming for real-time transcription and mai-voice-2.1 for text-to-speech. mai-transcribe-2-streaming transcribes 60 languages with high accuracy, delivering partial results in 100 milliseconds at an introductory price of $0.54 per hour of audio. mai-voice-2.1 can speak 23 languages with native accents and clone voices from short reference audio samples. both models have built-in safeguards to prevent misuse and are available on microsoft platforms and openrouter. these models can enhance voice agent interactions and customer service experiences.