Microsoft AI's text-to-speech model for keeping one voice across 23 languages, with voice cloning from short audio
Grok Voice Transcribe 2.0
New· 4xAI's second-generation transcription model for batch and live audio, with speaker separation and up to eight channels
About Grok Voice Transcribe 2.0
Grok Voice Transcribe 2.0, xAI's speech-to-text model, arrived on 18 September 2026; the company's documentation and website now use the name SpaceXAI. Developers reach it through the Speech to Text API under the identifier grok-voice-transcribe-2.0. xAI says its base is the foundation model for audio that also underlies Grok Voice, and that training used live, noisy, multilingual audio.
Files and live streams
Recorded files or URLs can be sent in 12 audio formats, up to 500 MB each, and live audio can go over WebSocket, where Opus packets suit clients short on bandwidth. The model detects the spoken language automatically.
Features for transcripts and voice agents
The feature list covers word-level timestamps, speaker diarization included at no extra charge, up to eight independently transcribed channels, and biasing toward as many as 100 key terms. It also removes filler words, applies smart turn detection for voice agents, and writes numbers and currencies in their written form in 25 languages. Pricing is unchanged from version 1.0, with separate rates for batch and streaming work and diarization, timestamps and key terms included.
The company's own accuracy claims
xAI's own testing, it says, shows version 2.0 roughly doubling the accuracy of 1.0; on a short-phrase set covering 19 languages, the word error rate dropped from 20.6 percent to 6.8 percent. It adds that, at launch, the model placed first for accuracy among 32 streaming models on the leaderboard kept by Artificial Analysis. MarkTechPost covered the release on 18 September and pointed out that these results come from the vendor. Atlassian is named as a launch customer for transcription in Loom.
Which version answers by default
The documentation disagrees with itself about the default: the speech-to-text page says 2.0 applies when no model is named, whereas the release notes say 1.0 stays the default. Since xAI plans to deprecate 1.0 within weeks, integrations should specify a model explicitly instead of relying on the default.
TrustList summary
xAI's speech-to-text model, Grok Voice Transcribe 2.0, was released on 18 September 2026 and can be called through the Speech to Text API using the model name grok-voice-transcribe-2.0. The company's documentation now gives its name as SpaceXAI.
Main features
- Batch and live transcription: recorded files, or links to them, up to 500 MB in 12 audio formats, plus live streams over WebSocket, with Opus packets supported for low-bandwidth clients.
- Speaker and timing detail: word-level timestamps, speaker diarisation and up to eight channels, each transcribed independently.
- Vocabulary control: biasing towards up to 100 key terms, with filler-word removal and smart turn detection for voice agents.
- Language handling: automatic language detection, and written-form formatting of numbers in 25 languages.
Customers and markets
Developers building voice agents and transcription features. Atlassian is named as a launch customer, using it for Loom transcription.
Key facts
- Accuracy claims: the vendor says version 2.0 is about twice as accurate as 1.0, with word error rate on a 19-language short-phrase set falling from 20.6% to 6.8%. These figures are vendor-reported.
- Default model: the documentation disagrees on which model is the default, so name the model explicitly.
- Deprecation: xAI says version 1.0 will be deprecated in the coming weeks.
- Lineage: the company says it draws on the audio foundation model behind Grok Voice.
Summary written by TrustList from the business's own published information, Nov 2026.
Benchmarks & AI stats
| Benchmark | Official | Community avg |
|---|---|---|
| No benchmark scores yet. Be the first to add one. | ||
“Official” values are editor-approved and feed the ranking. “Community avg” is the mean of member submissions (shown for transparency; it never affects the ranking until an editor approves a value).
Sign in to add a benchmark score for this model.
Grok Voice Transcribe 2.0 launches
Trust Score
An earned signal from verification, reviews, awards, transparency and engagement - the vendor can't buy it.
Updated 9/22/2026
Request a demo or quote from Grok Voice Transcribe 2.0
Reviews
Write the first review of Grok Voice Transcribe 2.0
Used it? Your experience helps other buyers decide.
Write a reviewQuestions & answers
No questions yet. Be the first to ask about Grok Voice Transcribe 2.0.
Grok Voice Transcribe 2.0 alternatives
Similar listings buyers compare against this one.
Aleph Alpha's open-weight English-German mixture-of-experts model for self-hosting, released under Apache 2.0 in October 2026
Google's open multimodal embedding model that maps text, images, audio and video into one vector space, under Apache 2.0
Open-source 2B decision model from AWS's Strands Agents project that picks options and returns scores with a reliability figure
Anthropic's smallest and fastest Claude 5.5 model, available on its own API and three cloud platforms
JetBrains' open 12B mixture-of-experts coding model with 2.5B parameters active per token under Apache 2.0
Open-weight translation model from Bilibili's Index team, covering 150 languages and following glossary and style instructions
Cantina Security's first open-weight security model, a post-trained GLM-5.3-Flash for vulnerability research and exploit checks
More on TrustList
Everything here links back to the same verified catalogue. Pick your next stop.
- Write a reviewHelp other buyers decide
- CompaniesAgencies, consultancies and IT service providers, ranked by verified reviews.
- ProductsSoftware and SaaS with pricing, features, integrations and alternatives.
- ArticlesGuides, comparisons and research from the editorial desk and the community.
- AwardsAnnual recognition decided by verified reviews and an independent jury.
- LaunchesNew products and releases, voted up by the community every day.
- RequestsBuyers describe what they need; vendors respond directly.
- PeopleReviewers, authors and makers with public profiles.
- ComparePut up to four listings side by side before you shortlist.