RemiFabre
commited on
Commit
·
b0cb5ad
1
Parent(s):
0ecbc6c
Minor README tweaks
Browse files
README.md
CHANGED
|
@@ -4,8 +4,8 @@ Conversational demo for the Reachy Mini robot combining OpenAI's realtime APIs,
|
|
| 4 |
|
| 5 |
## Overview
|
| 6 |
- Real-time audio conversation loop powered by the OpenAI realtime API and `fastrtc` for low-latency streaming.
|
| 7 |
-
- Layered motion system queues primary moves (dances, emotions, goto poses, breathing) while blending speech-reactive wobble and face-tracking.
|
| 8 |
- Camera capture can route to OpenAI multimodal vision or stay on-device with SmolVLM2 local analysis.
|
|
|
|
| 9 |
- Async tool dispatch integrates robot motion, camera capture, and optional facial-recognition helpers through a Gradio web UI with live transcripts.
|
| 10 |
|
| 11 |
## Installation
|
|
@@ -120,14 +120,14 @@ The app starts a Gradio UI served locally (http://127.0.0.1:7860/). When running
|
|
| 120 |
| `stop_dance` | Clear queued dances. | Core install only. |
|
| 121 |
| `play_emotion` | Play a recorded emotion clip via Hugging Face assets. | Needs `HF_TOKEN` for the recorded emotions dataset. |
|
| 122 |
| `stop_emotion` | Clear queued emotions. | Core install only. |
|
| 123 |
-
| `get_person_name` |
|
| 124 |
| `do_nothing` | Explicitly remain idle. | Core install only. |
|
| 125 |
|
| 126 |
## Development workflow
|
| 127 |
- Install the dev group extras: `uv sync --group dev` or `pip install -e .[dev]`.
|
| 128 |
- Run formatting and linting: `ruff check .`.
|
| 129 |
- Execute the test suite: `pytest`.
|
| 130 |
-
- When iterating on robot motions, keep the control loop responsive
|
| 131 |
|
| 132 |
## License
|
| 133 |
Apache 2.0
|
|
|
|
| 4 |
|
| 5 |
## Overview
|
| 6 |
- Real-time audio conversation loop powered by the OpenAI realtime API and `fastrtc` for low-latency streaming.
|
|
|
|
| 7 |
- Camera capture can route to OpenAI multimodal vision or stay on-device with SmolVLM2 local analysis.
|
| 8 |
+
- Layered motion system queues primary moves (dances, emotions, goto poses, breathing) while blending speech-reactive wobble and face-tracking.
|
| 9 |
- Async tool dispatch integrates robot motion, camera capture, and optional facial-recognition helpers through a Gradio web UI with live transcripts.
|
| 10 |
|
| 11 |
## Installation
|
|
|
|
| 120 |
| `stop_dance` | Clear queued dances. | Core install only. |
|
| 121 |
| `play_emotion` | Play a recorded emotion clip via Hugging Face assets. | Needs `HF_TOKEN` for the recorded emotions dataset. |
|
| 122 |
| `stop_emotion` | Clear queued emotions. | Core install only. |
|
| 123 |
+
| `get_person_name` | DeepFace-based recognition of the current person. | Disabled by default (`ENABLE_FACE_RECOGNITION=False`); requires `deepface` and a local face database. |
|
| 124 |
| `do_nothing` | Explicitly remain idle. | Core install only. |
|
| 125 |
|
| 126 |
## Development workflow
|
| 127 |
- Install the dev group extras: `uv sync --group dev` or `pip install -e .[dev]`.
|
| 128 |
- Run formatting and linting: `ruff check .`.
|
| 129 |
- Execute the test suite: `pytest`.
|
| 130 |
+
- When iterating on robot motions, keep the control loop responsive => offload blocking work using the helpers in `tools.py`.
|
| 131 |
|
| 132 |
## License
|
| 133 |
Apache 2.0
|