Python interrupt fix (#27)

Picovoice · Jan 2, 2025 · 131d266 · 131d266
1 parent ffc057c
commit 131d266
Show file tree

Hide file tree

Showing 8 changed files with 807 additions and 459 deletions.
diff --git a/recipes/llm-voice-assistant/python/README.md → .../llm-voice-assistant/python/cli/README.md b/recipes/llm-voice-assistant/python/README.md → .../llm-voice-assistant/python/cli/README.md
@@ -1,3 +1,7 @@
+# Cross-Platform LLM Voice Assistant CLI Demo
+
+A cross-platform voice assistant using Picovoice's Wake Word, STT, TTS and LLM technology with a text-based interface.
+
 ## See It In Action!
 
 [![LLM VA in Action](https://img.youtube.com/vi/06K_YtUr8mc/0.jpg)](https://www.youtube.com/watch?v=06K_YtUr8mc)
@@ -43,6 +47,17 @@ To see all available options, type the following:
 python main.py --help
 ```
 
+## Config File
+
+In addition to command line arguments a config file can be used to pass arguments to the demo. By default the demo looks for `config.json` in the same directory as `main.py` but an alternative path can be passed using the `--config` option. Below is an example config file.
+
+```json
+{
+    "access_key": "${ACCESS_KEY}",
+    "picollm_model_path": "${PICOLLM_MODEL_PATH}"
+}
+```
+
 ## Custom Wake Word
 
 The demo's default wake phrase is `Picovoice`. You can generate your custom (branded) wake word using Picovoice  Console by following [Porcupine Wake Word documentation (https://picovoice.ai/docs/porcupine/). Once you have the model trained, simply pass it to the demo