Fix unzip commands and instrument sources

README.md

@@ -42,17 +42,22 @@ The data is split into 8 zip archives (about 150 GB in total).
 
 | Archive | Extracts to |
 |---|---|
-| `truemuse_concepts.zip` | `concepts/` (extract with `-d data`, see below) |
+| `truemuse_concepts.zip` | `concepts/` (no `data/` prefix: extract with `-d data`) |
 | `truemuse_generated_audioldm2.zip` | `data/generated/audioldm2/` |
 | `truemuse_generated_mustango.zip` | `data/generated/mustango/` |
-| `truemuse_generated_stable_audio_{genre,instrument,melody,musician}.zip` | `data/generated/stable-audio/<task>/` |
+| `truemuse_generated_stable_audio_{genre,instrument,melody}.zip` | `data/generated/stable-audio/<task>/` |
+| `truemuse_generated_stable_audio_musician.zip` | `musician_{3,6}/` (no prefix: extract with `-d data/generated/stable-audio`) |
 | `truemuse_embeddings_jamendo.zip` | `data/embeddings/jamendo/` |
 
 ```bash
 hf download AnonymousAuthorsssss/TrueMuse --repo-type dataset --local-dir <dataset_root>
 cd <dataset_root>
-unzip -q truemuse_concepts.zip -d data
-for f in truemuse_generated_*.zip truemuse_embeddings_jamendo.zip; do unzip -q "$f"; done
+unzip -q truemuse_concepts.zip -d data                                                # no data/ prefix
+unzip -q truemuse_generated_stable_audio_musician.zip -d data/generated/stable-audio    # no data/ prefix
+for f in truemuse_generated_audioldm2.zip truemuse_generated_mustango.zip \
+         truemuse_generated_stable_audio_{genre,instrument,melody}.zip truemuse_embeddings_jamendo.zip; do
+    unzip -q "$f"
+done
 ```
 
 The instrument source audio is not included; rebuild it from YouTube with
@@ -104,8 +109,8 @@ source for instruments).
 
 Each fine-tuned model generates 4 waveforms per prompt from two prompt types: **contextual** prompts place the
 attribute in a richer musical scene, **style** prompts request it in a given genre (15 contextual + 28 style
-prompts; Stable Audio uses 15 + 8, and 15 + 28 for genre). Clips are 10 s long: 16 kHz for AudioLDM2 and
-Mustango, 44.1 kHz stereo for Stable Audio Open. The `metadata.csv` in each concept folder records the prompt
+prompts; Stable Audio uses 15 + 8, and 15 + 28 for genre). Clips are about 10 s long (10.24 s for Mustango):
+16 kHz mono for AudioLDM2 and Mustango, 44.1 kHz stereo for Stable Audio Open. The `metadata.csv` in each concept folder records the prompt
 of every clip.
 
 ### Distractor embeddings (`data/embeddings/jamendo/`)