Spacexai launches grok voice transcribe 2.0 with enhanced accuracy and multilingual support

SpaceXAI has launched Grok Voice Transcribe 2.0, a speech-to-text API claiming double the accuracy of its predecessor, Grok Voice Transcribe 1.0, at the same price of $0.10 per hour. The model is designed to handle challenging audio environments, including noisy phone lines, competing voices, and accents. It is available as a hosted API under the model ID grok-voice-transcribe-2.0 and supports real-time streaming and batch processing. The model is built on the audio foundation model behind Grok Voice, which already processes tens of thousands of customer-support calls daily and transcribes millions of hours of video narration. SpaceXAI reports that Grok Voice Transcribe 2.0 ranks first in accuracy among 32 streaming models on the Artificial Analysis leaderboard. The model also improves on 1.0 across four internal datasets, with a significant reduction in word error rate for short phrases, dropping from 20.6% to 6.8%. It supports 25 languages and automatically detects and handles mid-recording language switches. Atlassian Loom now uses the API to transcribe videos, citing improved accuracy over its existing solution.