diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..29ad255
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,27 @@
+.env
+
+# Python
+__pycache__/
+*.pyc
+.venv/
+venv/
+
+# Model weights / large artifacts
+*.pt
+*.ckpt
+*.bin
+*.safetensors
+models/
+weights/
+
+# Audio I/O scratch
+*.wav
+*.mp3
+*.flac
+output/
+outputs/
+
+# OS/editor
+.DS_Store
+.idea/
+.vscode/
diff --git a/README.md b/README.md
index a28aa1b..cf8c4a8 100644
--- a/README.md
+++ b/README.md
@@ -1,3 +1,15 @@
 # audio-demo
 
-Audio processing demos: WhisperX, Demucs, and TTS technology proofs-of-concept.
\ No newline at end of file
+Audio processing demos: WhisperX, Demucs, and TTS technology proofs-of-concept.
+
+## Structure
+
+- `whisperx/` — speech-to-text with word-level alignment and diarization
+- `demucs/` — source separation (vocals/instrumentals stem splitting)
+- `tts/` — text-to-speech service demos
+
+Each subdirectory has its own README with setup and run instructions.
+
+## Requirements
+
+Some demos are GPU-accelerated and are best run on a CUDA-capable machine (e.g. an RTX 3090).
diff --git a/demucs/README.md b/demucs/README.md
new file mode 100644
index 0000000..a1f3f13
--- /dev/null
+++ b/demucs/README.md
@@ -0,0 +1,3 @@
+# demucs
+
+Source separation demos using [Demucs](https://github.com/facebookresearch/demucs) — splitting audio into stems (vocals, drums, bass, other).
diff --git a/tts/README.md b/tts/README.md
new file mode 100644
index 0000000..0451ccb
--- /dev/null
+++ b/tts/README.md
@@ -0,0 +1,3 @@
+# tts
+
+Text-to-speech service demos and proofs-of-concept.