Add talking_Xs actions and update speaking guidelines#2618
Merged
Conversation
Add a Speaking Action Selection section to greeting_conversation.json5 and unitree_g1_conversation.json5 that instructs selecting a single talking_Xs action based on estimated spoken duration (talking_2s..talking_20s) and to prefer gestures when appropriate. Update the Unitree G1 system prompt to include the new talking_* actions and change ROTATE_HAND -> ROTATE_HANDS. Update ArmAction EnumValues in plugins/actions/unitree/g1/arm/zenoh.go to expose talking_2s..talking_20s and remove the old speak_action entries so runtime action enums match the new guidance.
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
Contributor
There was a problem hiding this comment.
Pull request overview
This pull request updates Unitree G1 conversation behavior by replacing generic speaking actions with duration-based talking_Xs actions, and adjusts guidance so arm/body motions better match spoken response length.
Changes:
- Added duration-based talking actions (
talking_2s…talking_20s) to the Unitree G1 arm action enum. - Updated Unitree G1 conversation prompt guidelines to select a single talking action based on estimated speech duration, and clarified gesture vs. talking usage.
- Removed legacy generic speaking actions from the Unitree G1 conversation configuration prompt.
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 4 comments.
| File | Description |
|---|---|
| plugins/actions/unitree/g1/arm/zenoh.go | Extends the arm action enum with talking_Xs actions (and removes legacy speaking actions). |
| config/unitree_g1_conversation.json5 | Updates the Unitree G1 conversation system prompt to use duration-based talking actions and clarifies gesture usage. |
| config/greeting_conversation.json5 | Adds duration-based “Speaking Action Selection” guidance to greeting/conversation prompts. |
Comment on lines
+37
to
+41
| "talking_2s", | ||
| "talking_4s", | ||
| "talking_6s", | ||
| "talking_8s", | ||
| "talking_10s", |
Comment on lines
+88
to
+92
| Speaking Action Selection:\n\ | ||
| - When generating a spoken response, estimate how long it will take to speak naturally.\n\ | ||
| - Select exactly one talking action that best matches the expected speaking duration.\n\ | ||
| - Use talking_2s for responses up to 2 seconds.\n\ | ||
| - Use talking_4s for responses between 2 and 4 seconds.\n\ |
Comment on lines
+231
to
+235
| Speaking Action Selection:\n\ | ||
| - When generating a spoken response, estimate how long it will take to speak naturally.\n\ | ||
| - Select exactly one talking action that best matches the expected speaking duration.\n\ | ||
| - Use talking_2s for responses up to 2 seconds.\n\ | ||
| - Use talking_4s for responses between 2 and 4 seconds.\n\ |
Comment on lines
+37
to
+39
| "talking_2s", | ||
| "talking_4s", | ||
| "talking_6s", |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This pull request updates the behavior guidelines and action definitions for the Unitree G1 robot to introduce a new, duration-based set of talking actions for more natural and expressive speech coordination. It also removes the old generic speaking actions and clarifies gesture usage. The changes affect both configuration files and the action enumeration in the code.
Key changes:
Speaking Action Overhaul
talking_2s,talking_4s, ...,talking_20s) that correspond to different speaking durations, and updated the guidelines to select the appropriate action based on the expected length of the spoken response. [1] [2] [3] [4] [5]speak_action,speak_action_extended) from both the configuration and the code, ensuring only the new duration-based actions are available. [1] [2] [3]Gesture and Action Clarifications
rotate_handtorotate_hands) and clarified when to use gestures versus talking actions in the guidelines. [1] [2]STAND_STILLshould be used only when idle or listening.These changes make the robot's behavior more expressive and context-appropriate during conversations.