Thank you for this excellent work and for making the code publicly available! I'm working on a real-time speech-driven animation application and would like to use a 1-second time window (25 frames at 25fps) instead of the default 4-second window to minimise latency.
Could you please release the pre-trained weights for the 25-frame window configuration and designate the needed modifications in the code (e.g. config.json file? Thank you very much!
Thank you for this excellent work and for making the code publicly available! I'm working on a real-time speech-driven animation application and would like to use a 1-second time window (25 frames at 25fps) instead of the default 4-second window to minimise latency.
Could you please release the pre-trained weights for the 25-frame window configuration and designate the needed modifications in the code (e.g. config.json file? Thank you very much!