Update linux-voice-assistant-2mic-install.md
This commit is contained in:
@@ -27,7 +27,17 @@ sudo reboot
|
||||
```
|
||||
|
||||
|
||||
## 2. Get the code
|
||||
## 2. Install ReSpeaker drivers
|
||||
|
||||
```bash
|
||||
cd linux-voice-assistant/respeaker2mic/
|
||||
chmod +x ./install-respeaker-drivers.sh
|
||||
sudo ./install-respeaker-drivers.sh
|
||||
sudo reboot
|
||||
```
|
||||
|
||||
|
||||
## 3. Get the code
|
||||
|
||||
```bash
|
||||
git clone https://github.com/imonlinux/linux-voice-assistant.git
|
||||
@@ -35,7 +45,7 @@ git clone https://github.com/rhasspy/wyoming-openwakeword.git
|
||||
```
|
||||
|
||||
|
||||
## 3. Wyoming OpenWakeWord (OWW)
|
||||
## 4. Wyoming OpenWakeWord (OWW)
|
||||
|
||||
```bash
|
||||
cp ./linux-voice-assistant/wyoming-openwakeword/requirements.txt ./wyoming-openwakeword/requirements.txt
|
||||
@@ -44,7 +54,7 @@ script/setup
|
||||
```
|
||||
|
||||
|
||||
## 4. Configure audio devices
|
||||
## 5. Configure audio devices
|
||||
|
||||
Find your sound device names and update the linux-voice-assistant/service/linux-voice-assistant.service file to match sound card details:
|
||||
```bash
|
||||
@@ -53,7 +63,7 @@ aplay -l
|
||||
```
|
||||
|
||||
|
||||
## 5. Systemd services
|
||||
## 6. Systemd services
|
||||
|
||||
Copy the 2 service files into /etc/systemd/system/:
|
||||
```bash
|
||||
@@ -68,7 +78,7 @@ sudo systemctl status linux-voice-assistant wyoming-openwakeword --no-pager -l
|
||||
```
|
||||
|
||||
|
||||
## 6. Connect to Home Assistant
|
||||
## 7. Connect to Home Assistant
|
||||
|
||||
1. In Home Assistant, go to "Settings" -> "Device & services"
|
||||
2. Click the "Add integration" button
|
||||
@@ -77,14 +87,14 @@ sudo systemctl status linux-voice-assistant wyoming-openwakeword --no-pager -l
|
||||
5. Click "Submit"
|
||||
|
||||
|
||||
## 7. Verification
|
||||
## 8. Verification
|
||||
|
||||
- Expect logs like `Connected to Home Assistant`
|
||||
- Look for `[OWW] Detection: name=...` followed by re-arming/cycling
|
||||
- Ask: *“What time is it?”* and confirm TTS reply
|
||||
|
||||
|
||||
## 8. Change OWW detection model
|
||||
## 9. Change OWW detection model
|
||||
|
||||
Edit the linux-voice-assistant.service file and change the OWW configuration argument for --wake-word-name.
|
||||
Project OWW models include:
|
||||
@@ -165,7 +175,7 @@ sudo systemctl restart linux-voice-assistant.service
|
||||
sudo systemctl status linux-voice-assistant wyoming-openwakeword --no-pager -l
|
||||
```
|
||||
|
||||
## 9. Revert to MicroWakeWord
|
||||
## 10. Revert to MicroWakeWord
|
||||
|
||||
Edit the linux-voice-assistant.service file and remove the OWW configuration arguments:
|
||||
|
||||
|
||||
Reference in New Issue
Block a user