Update README.md

README.md

@@ -51,7 +51,28 @@ llama-cli --hf-repo hellork/calme-2.1-qwen2-7b-Q4_K_M-GGUF --hf-file calme-2.1-q
 llama-server --hf-repo hellork/calme-2.1-qwen2-7b-Q4_K_M-GGUF --hf-file calme-2.1-qwen2-7b-q4_k_m.gguf -c 2048
 ```
 
-Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
+### The Ship's Computer:
+
+[whisper_dictation](https://github.com/themanyone/whisper_dictation)
+
+Interact with this model by speaking to it. Lean, fast, & private, networked speech to text, AI images, multi-modal voice chat, control apps, webcam, and sound with less than 4GiB of VRAM.
+
+```bash
+git clone -b main --single-branch https://github.com/themanyone/whisper_dictation.git
+pip install -r whisper_dictation/requirements.txt
+
+git clone https://github.com/ggerganov/whisper.cpp
+cd whisper.cpp
+GGML_CUDA=1 make -j # assuming CUDA is available. see docs
+ln -s server ~/.local/bin/whisper_cpp_server # (just put it somewhere in $PATH)
+
+whisper_cpp_server -l en -m models/ggml-tiny.en.bin --port 7777
+cd whisper_dictation
+./whisper_cpp_client.py
+```
+See [the docs](https://github.com/themanyone/whisper_dictation) for tips on integrating with llama.cpp server, enabling the computer to talk back, draw AI images, carry out voice commands, and other features.
+
+### Install Llama.cpp via git:
 
 Step 1: Clone llama.cpp from GitHub.
 ```