Turn-based chat interface
Stack
- LLM: ChatGPT
- Speech-to-text: Whisper
- Text-to-speech: Amazon Polly
- Actor Model framework: Actix
- Audio playback library: rodio
- Audio input library: cpal
-
Download Protobuf.
brew install protobuf
Protobuf is needed for gRPC with the vector database.
-
Download libtorch v1.13.1 for macOS CPU from here. Then unzip it.
-
Set the following env variables:
export LIBTORCH=<path-to-unzipped-folder> export DYLD_FALLBACK_LIBRARY_PATH=${LIBTORCH}/lib # for macOS
For other platforms, see here to specify dynamic library path.
-
Run the Qdrant container.
docker run -d -p 6333:6333 -p 6334:6334 \ -v $(pwd)/qdrant_storage:/qdrant/storage \ qdrant/qdrant
-
Run
cargo run
VS Code
"rust-analyzer.cargo.extraEnv": {
"LIBTORCH": "/root/miniconda/lib/python3.10/site-packages/torch",
"LD_LIBRARY_PATH": "/root/miniconda/lib/python3.10/site-packages/torch/lib",
"DYLD_FALLBACK_LIBRARY_PATH": "/root/miniconda/lib/python3.10/site-packages/torch/lib",
"LIBTORCH_CXX11_ABI": "0"
}