SpaceXAI Releases Grok Voice Transcribe 2.0: A Speech-to-Text API Claiming 2x Accuracy Over 1.0 at $0.10 per Hour
SpaceXAI released Grok Voice Transcribe 2.0, a speech-to-text API claiming 2x accuracy over 1.0 at unchanged $0.10-$0.20 per hour pricing.
SpaceXAI launched Grok Voice Transcribe 2.0, a hosted speech-to-text model (model ID grok-voice-transcribe-2.0) built on the audio foundation model behind Grok Voice, available via batch and WebSocket streaming endpoints. The company reports first place among 32 streaming models on the Artificial Analysis AA-WER Streaming leaderboard and says short-phrase word error rate across 19 languages dropped from 20.6% to 6.8%. The API includes word-level timestamps, speaker diarization, up to 8-channel transcription, and key-term biasing, with batch at $0.10 per audio hour and streaming at $0.20. Atlassian Loom has adopted the model for video transcription, feeding transcripts into Cursor for code generation.
- Claims 2x accuracy over Grok Voice Transcribe 1.0 at unchanged pricing ($0.10/hr batch, $0.20/hr streaming).
- Vendor-reported rank 1 among 32 streaming models on Artificial Analysis AA-WER Streaming benchmark.
- Short-phrase WER across 19 languages fell from 20.6% to 6.8%, roughly 67% fewer errors.
- API-only with no open weights; supports diarization, 8 channels, word timestamps, and key-term biasing.
- Atlassian Loom adopted the model for transcription, piping transcripts into Cursor for coding.
- AI models
- grok-voice-transcribe-2.0
From the original article
4 min read at marktechpost.com
SpaceXAI has released Grok Voice Transcribe 2. 0, its newest speech-to-text model for batch and streaming audio. The company says it is twice as accurate as version 1.
By Michal Sutter · marktechpost.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.