Document all three ONNX files

README.md

@@ -9,37 +9,38 @@ library_name: openwakeword
 pipeline_tag: audio-classification
 ---
 
-# BMO openWakeWord model
+# BMO openWakeWord models
 
-Custom [openWakeWord](https://github.com/dscripka/openWakeWord) model for the wake phrase **“Hey BMO”** (trained as `hey beemo`).
+Custom [openWakeWord](https://github.com/dscripka/openWakeWord) pack for the wake phrase **“Hey BMO”** (trained as `hey beemo`).
 
 ## Files
 
 | File | Description |
 |------|-------------|
-| `hey_beemo.onnx` | Wake-word classifier head (ONNX) |
+| `hey_beemo.onnx` | Wake-word classifier (“Hey BMO”) |
+| `melspectrogram.onnx` | Shared openWakeWord mel feature model |
+| `embedding_model.onnx` | Shared openWakeWord speech embedding model |
 
-Also requires the shared openWakeWord feature models at runtime:
+All three are required at runtime.
 
-- `melspectrogram.onnx`
-- `embedding_model.onnx`
+Shared feature models come from [openWakeWord-cpp](https://github.com/rhasspy/openWakeWord-cpp) / [openWakeWord](https://github.com/dscripka/openWakeWord).
 
-(from [openWakeWord](https://github.com/dscripka/openWakeWord) / [openWakeWord-cpp](https://github.com/rhasspy/openWakeWord-cpp))
+## Download
 
-## Usage with local-llm
-
-```json
-"model": {
-  "path": "../models/ww/hey_beemo.onnx"
-}
+```bash
+hf download amrmantawi/bmo-openwakeword --local-dir models/ww
 ```
 
-Or download from this repo:
+## local-llm config
 
-```bash
-hf download amrmantawi/bmo-openwakeword hey_beemo.onnx --local-dir models/ww
+```json
+"openwakeword": {
+  "melspectrogram": { "path": "../models/ww/melspectrogram.onnx" },
+  "embedding": { "path": "../models/ww/embedding_model.onnx" },
+  "model": { "path": "../models/ww/hey_beemo.onnx" }
+}
 ```
 
 ## Training
 
-Trained with the openWakeWord Google Colab automatic training notebook on synthetic speech for the phrase **hey beemo**.
+`hey_beemo.onnx` was trained with the openWakeWord Google Colab notebook on synthetic speech for **hey beemo**.