Skip to content

Commit d288590

Browse files
authored
Merge pull request #917 from pipecat-ai/docs/pr-4785
docs: update RTVI documentation for VAD user speaking events
2 parents 0ad1fd7 + 73426c5 commit d288590

2 files changed

Lines changed: 30 additions & 0 deletions

File tree

api-reference/server/rtvi/rtvi-observer.mdx

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -59,6 +59,10 @@ worker = PipelineWorker(
5959
Indicates if the user's started/stopped speaking messages should be sent.
6060
</ParamField>
6161

62+
<ParamField path="vad_user_speaking_enabled" default="False">
63+
Indicates if raw VAD user started/stopped speaking messages should be sent. These reflect the VAD signal directly, independent of turn finalization (unlike `user_speaking_enabled`, which a turn strategy may gate or defer).
64+
</ParamField>
65+
6266
<ParamField path="user_mute_enabled" default="True">
6367
Indicates if user mute started/stopped messages (`user-mute-started`,
6468
`user-mute-stopped`) should be sent.
@@ -188,6 +192,8 @@ The observer maps Pipecat's internal frames to RTVI protocol messages:
188192
| **Speech Events** |
189193
| `UserStartedSpeakingFrame` | `RTVIUserStartedSpeakingMessage` |
190194
| `UserStoppedSpeakingFrame` | `RTVIUserStoppedSpeakingMessage` |
195+
| `VADUserStartedSpeakingFrame` | `VADUserStartedSpeakingMessage` |
196+
| `VADUserStoppedSpeakingFrame` | `VADUserStoppedSpeakingMessage` |
191197
| `BotStartedSpeakingFrame` | `RTVIBotStartedSpeakingMessage` |
192198
| `BotStoppedSpeakingFrame` | `RTVIBotStoppedSpeakingMessage` |
193199
| **User Mute** |

client/rtvi-standard.mdx

Lines changed: 24 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -167,6 +167,30 @@ Emitted when the user stops speaking
167167
- **type**: `'user-stopped-speaking'`
168168
- **data**: None
169169

170+
#### vad-user-started-speaking 🤖
171+
172+
<Note>
173+
Disabled by default; enable with
174+
`RTVIObserverParams(vad_user_speaking_enabled=True)`.
175+
</Note>
176+
177+
Emitted when the VAD (Voice Activity Detection) detects the user started speaking. This is the raw VAD signal, emitted independently of turn finalization (unlike `user-started-speaking`, which a turn strategy may gate or defer).
178+
179+
- **type**: `'vad-user-started-speaking'`
180+
- **data**: None
181+
182+
#### vad-user-stopped-speaking 🤖
183+
184+
<Note>
185+
Disabled by default; enable with
186+
`RTVIObserverParams(vad_user_speaking_enabled=True)`.
187+
</Note>
188+
189+
Emitted when the VAD (Voice Activity Detection) detects the user stopped speaking. This is the raw VAD signal, emitted independently of turn finalization (unlike `user-stopped-speaking`, which a turn strategy may gate or defer).
190+
191+
- **type**: `'vad-user-stopped-speaking'`
192+
- **data**: None
193+
170194
#### bot-started-speaking 🤖
171195

172196
Emitted when the bot begins speaking

0 commit comments

Comments
 (0)