You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: rest/README.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -2,7 +2,7 @@
2
2
3
3
Turn-by-turn REST approach: click **Next Turn**, wait for a full turn to complete, hear the audio. Each turn is a complete HTTP request/response cycle. This is the right starting point for understanding how voice agent pieces fit together before adding streaming complexity.
4
4
5
-
> **She Ships! — February 25, 2026** — This REST demo was the basis for the live demo at that talk ([slide deck PDF](../assets/SheShips_When_Voice_Agents_Roll_Initiative.pdf)). The repo has evolved since then:
5
+
> **She Ships! — February 25, 2026** — This REST demo was the basis for the [live demo](https://youtu.be/-mp9beFRJ1Y) at that talk ([slide deck PDF](../assets/SheShips_When_Voice_Agents_Roll_Initiative.pdf)). The repo has evolved since then:
6
6
> - The decorative dice animation became a real d20 mechanic — `workflow.random().randint(1, 20)` inside the Temporal workflow handler keeps rolls deterministic and replay-safe
7
7
> - A DM narration activity (`claude-haiku-4-5` or `gpt-4o-mini`) generates one sentence describing the roll outcome, displayed at the top of the *next* turn so it lands as a beat between characters after the audio has played
0 commit comments