Google launches Gemini 3.5 Transcribe and Play Books import in Gemini Notebook
Google announced two AI updates: a new speech-to-text model, Gemini 3.5 Transcribe, and a feature that lets users import purchased Play Books into Gemini Notebook.
According to MarkTechPost, Gemini 3.5 Transcribe is a managed API service, not an open-weight model. It consists of two endpoints: gemini-3.5-transcribe for pre-recorded files through the Interactions API, and gemini-3.5-transcribe-live for real-time bidirectional streaming through the Live API. Google reports average word error rates of 4.0% for streaming and 2.6% for non-streaming, based on measurements by Artificial Analysis. The company also claims the time to final transcription is 70% faster than its predecessor, Chirp 3.
The service is available through three tiers: a free tier for solo developers via Google AI Studio, a paid tier with higher rate limits and a guarantee that content is not used to improve Google’s products, and an enterprise tier through the Gemini Enterprise Agent Platform with provisioned throughput, compliance controls, and volume discounts. Both developer and enterprise tracks are in public preview.
The model supports more than 85 languages with automatic detection and mid-sentence code-switching, according to the report. The Live API delivers sub-second continuous transcription with interim partials, but sessions are capped at 10 minutes, and speaker diarization and word-level timestamps are not supported. The Interactions API handles longer audio, up to one hour (30 minutes when diarization or word timestamps are enabled), and adds diarization, word timestamps, and custom vocabulary biasing of up to 1,000 terms.
Both endpoints offer two transcription modes: “verbatim” returns everything, including fillers and repetitions, while “smart” removes disfluencies and resolves self-corrections. Google gives the example of “Um, so for the meeting, I think we should, uh, invite Alice and, wait no, Bob and Carol.” In verbatim mode all disfluencies remain; in smart mode it becomes “For the meeting, I think we should invite Bob and Carol.” Smart mode cannot be combined with word timestamps or diarization.
Pricing is listed as about $0.005 per minute for batch transcription and $0.009 per minute for live transcription. The Live API is integrated with LiveKit, Pipecat, Agora, Fishjam, Vercel, and Vision Agents. Consumer products already using the model include Rambler on Android, the Gemini app on macOS, and Google Antigravity, with Chrome support said to be coming soon.
According to Android Authority, Google now allows users to import titles they own from Google Play Books into Gemini Notebook through a new “Expert Intelligence” option. The feature is the result of a partnership with authors to make certain paid books available in the tool. Users can combine insights from multiple titles into a single notebook and use features such as Audio Overviews and Infographics.
The announcement follows ongoing controversy over using copyrighted books for AI training. Google is now formally opening a channel for users to engage with copyrighted books within Gemini Notebook, under a partnership with authors.