Cache for whisper model

.dockerignore

@@ -0,0 +1,6 @@
+*.txt
+*.json
+*.log
+*.md
+*.sh
+model

.gitignore

@@ -196,4 +196,5 @@ pyrightconfig.json
 .history
 .ionide
 
-# End of https://www.toptal.com/developers/gitignore/api/visualstudiocode,python
+# End of https://www.toptal.com/developers/gitignore/api/visualstudiocode,python
+model/base.pt

README.md

@@ -48,17 +48,21 @@ You can run the Docker container with the following command, mapping the necessa
 
 ```bash
 docker run -it --rm \
-  -v /path/to/your/local/files:/app \
+  -v ./:/app \
+  -v ./model:/root/.cache/whisper \
   -e INPUT_FILE="file.mp4" \
   -e PROMPT_FILE="meeting.md" \
   -e CONFIG_FILE="config.yml" \
   echomind
 ```
 
 This command:
-- Mounts the directory containing your files to the `/app` directory inside the Docker container.
+- Mounts the root of the repo to the `/app` directory inside the Docker container.
+- Mounts the `model` directory to the download location of the whisper model to avoid repeat downloads
 - Sets environment variables for the input file, prompt file, and config file.
 
+If your Ollama installation is on the same host, you will want to include `--network host` to the command in order to run the container with the ability to interact with the host's loopback address.
+
 ### Output Files
 
 The script will generate the following output files based on the input filename:

run.sh

@@ -1,5 +1,5 @@
 #!/bin/bash
- 
+
  docker build -t echomind .
 
 # Check if correct number of arguments are provided
@@ -16,6 +16,7 @@ CONFIG_FILE=$3
 # Run the Docker command with the provided arguments
 docker run -it --rm \
   -v "$(pwd):/app" \
+  -v "$(pwd)/model:/root/.cache/whisper" \
   -e INPUT_FILE="$INPUT_FILE" \
   -e PROMPT_FILE="$PROMPT_FILE" \
   -e CONFIG_FILE="$CONFIG_FILE" \