feat(voice): implement real-time voice mode with cloud and local backends (#24174)

This commit is contained in:
Abhijit Balaji
2026-04-24 14:29:38 -07:00
committed by GitHub
parent 048bf6e514
commit 2e0641c83b
40 changed files with 2244 additions and 43 deletions
+26
View File
@@ -1691,6 +1691,32 @@ their corresponding top-level category object in your `settings.json` file.
- **Default:** `false`
- **Requires restart:** Yes
- **`experimental.voiceMode`** (boolean):
- **Description:** Enable experimental voice dictation and commands (/voice,
/voice model).
- **Default:** `false`
- **`experimental.voice.activationMode`** (enum):
- **Description:** How to trigger voice recording with the Space key.
- **Default:** `"push-to-talk"`
- **Values:** `"push-to-talk"`, `"toggle"`
- **`experimental.voice.backend`** (enum):
- **Description:** The backend to use for voice transcription.
- **Default:** `"gemini-live"`
- **Values:** `"gemini-live"`, `"whisper"`
- **`experimental.voice.whisperModel`** (enum):
- **Description:** The Whisper model to use for local transcription.
- **Default:** `"ggml-base.en.bin"`
- **Values:** `"ggml-tiny.en.bin"`, `"ggml-base.en.bin"`,
`"ggml-large-v3-turbo-q5_0.bin"`, `"ggml-large-v3-turbo-q8_0.bin"`
- **`experimental.voice.stopGracePeriodMs`** (number):
- **Description:** How long to wait for final transcription after stopping
recording.
- **Default:** `1000`
- **`experimental.adk.agentSessionNoninteractiveEnabled`** (boolean):
- **Description:** Enable non-interactive agent sessions.
- **Default:** `false`
+1
View File
@@ -115,6 +115,7 @@ available combinations.
| `app.restart` | Restart the application. | `R`<br />`Shift+R` |
| `app.suspend` | Suspend the CLI and move it to the background. | `Ctrl+Z` |
| `app.showShellUnfocusWarning` | Show warning when trying to move focus away from shell input. | `Tab` |
| `app.voiceModePTT` | Hold to speak in Voice Mode. | `Space` |
#### Background Shell Controls