pec-dss 0.1.0__py3-none-any.whl → 0.1.1__py3-none-any.whl

pec_dss/__init__.py

@@ -5,4 +5,4 @@ An advanced audio analysis system that identifies paralinguistic vocal events
 and attributes them to specific speakers through sophisticated speaker diarization.
 """
 
-__version__ = "0.1.0"
+__version__ = "0.1.1"

{pec_dss-0.1.0.dist-info → pec_dss-0.1.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pec-dss
-Version: 0.1.0
+Version: 0.1.1
 Summary: Paralinguistic Event Classification from Diarized Speaker Segments
 Author-email: Hyunwoo Kim <hwk06023@gmail.com>
 License: GPL-3.0-only
@@ -118,6 +118,56 @@ for speaker, events in results.items():
     print(f"Speaker {speaker} has {len(events)} attributed events")
 ```
 
+### CLI Usage
+
+```bash
+pec-dss --speakers-dir ./speakers --unidentified-dir ./events --output-dir ./results
+```
+
+## 📁 Directory Structure
+
+PEC-DSS expects a specific directory structure for processing audio files:
+
+### Speaker Reference Structure
+
+```
+speakers_directory/
+   ├── speaker_A/       # Each speaker's name becomes their ID
+   │   ├── audio1.wav   # Reference voice samples for this speaker
+   │   ├── audio2.wav
+   │   └── ...
+   ├── speaker_B/
+   │   ├── audio1.wav
+   │   └── ...
+   └── speaker_C/
+       ├── audio1.wav
+       └── ...
+```
+
+### Unidentified Audio Structure
+
+```
+unidentified_directory/
+   ├── laugh1.wav      # Non-linguistic vocal events to be classified
+   ├── giggle1.wav
+   └── ...
+```
+
+### Output Structure (After Processing)
+
+```
+output_directory/
+   ├── results.json           # JSON file with all results
+   ├── speaker_A/             # Files assigned to each speaker
+   │   ├── 0_laugh1.wav
+   │   └── ...
+   ├── speaker_B/
+   │   ├── 0_giggle1.wav
+   │   └── ...
+   └── unknown/               # Files below similarity threshold (if any)
+       └── ...
+```
+
 ## 🧩 System Architecture
 
 PEC-DSS consists of the following components:
@@ -138,6 +188,8 @@ The system can identify various paralinguistic events including:
 * Coughing
 * Other non-verbal vocal expressions
 
+**Note:** PEC-DSS does not automatically classify these event types. It only determines which speaker produced the audio event.
+
 ## 🚀 Future Developments
 
 * 🧠 Integration with more audio encoder models

pec_dss-0.1.1.dist-info/RECORD

@@ -0,0 +1,9 @@
+pec_dss/__init__.py,sha256=QyuqbTvRCIgEd7NbGi7I521W5y0wYNvEVM11_Q6N79A,271
+pec_dss/__main__.py,sha256=Ul7rrlyhsspPqAsbsJul1kzckUqBlx1irsoWwPJBoz0,144
+pec_dss/cli.py,sha256=bDopqySnvDeQH7ZhdDlkDg5sxzmFV5gYpx-tWm6Dcyk,3952
+pec_dss-0.1.1.dist-info/licenses/LICENSE,sha256=ixuiBLtpoK3iv89l7ylKkg9rs2GzF9ukPH7ynZYzK5s,35148
+pec_dss-0.1.1.dist-info/METADATA,sha256=Vz3gYZRTS9IhLBS2CgI51abLJfVn2R9OPXMsC_TRHfI,6448
+pec_dss-0.1.1.dist-info/WHEEL,sha256=pxyMxgL8-pra_rKaQ4drOZAegBVuX-G_4nRHjjgWbmo,91
+pec_dss-0.1.1.dist-info/entry_points.txt,sha256=birA2tqMJICs1CXYBcEpLKiK3HgKvH0Z494fgHgwjXQ,45
+pec_dss-0.1.1.dist-info/top_level.txt,sha256=vdoLeQ6vlHAHGdWl9DrCP09WTySHt1mfiqji8gRvSq4,8
+pec_dss-0.1.1.dist-info/RECORD,,

pec_dss-0.1.0.dist-info/RECORD

@@ -1,9 +0,0 @@
-pec_dss/__init__.py,sha256=jibqnYamtHCJ8C9VcoxSKpgTMc114TuHFSCpctj5ctk,271
-pec_dss/__main__.py,sha256=Ul7rrlyhsspPqAsbsJul1kzckUqBlx1irsoWwPJBoz0,144
-pec_dss/cli.py,sha256=bDopqySnvDeQH7ZhdDlkDg5sxzmFV5gYpx-tWm6Dcyk,3952
-pec_dss-0.1.0.dist-info/licenses/LICENSE,sha256=ixuiBLtpoK3iv89l7ylKkg9rs2GzF9ukPH7ynZYzK5s,35148
-pec_dss-0.1.0.dist-info/METADATA,sha256=jGLZdXkQwLuusf0VmV0wpmnp0vBykC7hAsUpFePFk4M,5069
-pec_dss-0.1.0.dist-info/WHEEL,sha256=pxyMxgL8-pra_rKaQ4drOZAegBVuX-G_4nRHjjgWbmo,91
-pec_dss-0.1.0.dist-info/entry_points.txt,sha256=birA2tqMJICs1CXYBcEpLKiK3HgKvH0Z494fgHgwjXQ,45
-pec_dss-0.1.0.dist-info/top_level.txt,sha256=vdoLeQ6vlHAHGdWl9DrCP09WTySHt1mfiqji8gRvSq4,8
-pec_dss-0.1.0.dist-info/RECORD,,