Whisper Transcription Locally: Where Realtime Ends on RK3588
Whisper is entirely usable on RK3588, but offline batch and streaming have very different expectations.
Batch works; realtime takes conditions
- Offline batch: the small model transcribes one hour of audio in 15-25 minutes; faster-whisper INT8 goes faster still;
- Streaming: needs VAD segmentation plus chunked inference with 3-8 s latency — fine for subtitles, not interpretation;
- Memory: small INT8 is about 1 GB; medium wants 8 GB-class devices.
For Chinese, try small with an initial_prompt of domain terms first — a free and noticeable accuracy boost.
Related products: RK3588 Mini PC, RP01A dev board. Datasheets, firmware and SDKs are free at the Hugsun Documentation Center; for selection advice or volume pricing, talk to us.
