Andy Wood, head of video at Spellbook, has built a voice control system for the Sony FX3/FX6 using a Raspberry Pi Zero, Sony’s camera SDK, and the Vosk open-source speech recognition API.
This was a side project that Andy did using Claude code to write a little piece of software that would allow him to use voice commands to trigger certain actions on the Sony FX6. Andy did this because he thought if he ever had to run a two-camera setup, that would be convenient for leveraging voice commands.
Sony FX3/ FX6 with voice recognition that I built on a raspberry pi zero running Sony SDK, and Vosk open source voice api. “FX6 roll!” https://t.co/XumG2eod6B
— Andy Wood (@heyimandy) October 10, 2026
You can say, “FX6 roll!” and the camera will start recording.
Technically, the system utilizes a Raspberry Pi Zero 2W running the official Sony SDK and Vosk, an open-source voice recognition API. Andy then added a small USB mic and printed an enclosure for the whole unit.
It is then all powered via a V-mount battery, which powers both the Raspberry Pi and the FX6.
The product is still in development, and Andy is still working on further testing and capabilities, but as of right now, this is what it can do:
- “FX6 roll” starts recording
- “FX6 cut” stops recording
- “FX6 mark” logs a timestamp to a local saved file
- “FX6 iris up” opens the aperture a stop for native lenses
- “FX6 iris down” closes the aperture a stop for native lenses
- “FX6 filter up” ND filter up a step
- “FX6 filter down” ND filter down a step
Currently, this is still a DIY project, but if it garners enough interest, it could turn into a product that you may eventually be able to buy.
