Hi EchoAvatar team,
Thank you for releasing the real-time deployment code and checkpoints. The streaming demo and the tool-call control interface are very useful for downstream research and prototyping.
I noticed that the README release plan lists Training code as a planned item. Do you have any rough expectation for when it may become available?
Even if there is no firm date yet, a high-level note on whether the release is expected to include the motion tokenizer / predictor training pipeline would help downstream users plan reproduction and adaptation work more responsibly.
No pressure at all if the timeline is still uncertain. Thanks again for the great work.
Hi EchoAvatar team,
Thank you for releasing the real-time deployment code and checkpoints. The streaming demo and the tool-call control interface are very useful for downstream research and prototyping.
I noticed that the README release plan lists Training code as a planned item. Do you have any rough expectation for when it may become available?
Even if there is no firm date yet, a high-level note on whether the release is expected to include the motion tokenizer / predictor training pipeline would help downstream users plan reproduction and adaptation work more responsibly.
No pressure at all if the timeline is still uncertain. Thanks again for the great work.