Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-4.0
|
| 3 |
+
tags:
|
| 4 |
+
- speech
|
| 5 |
+
- keyword-spotting
|
| 6 |
+
- evaluation
|
| 7 |
+
- noise-robustness
|
| 8 |
+
language:
|
| 9 |
+
- en
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# SpeechGuard Evaluation Set
|
| 13 |
+
|
| 14 |
+
Evaluation manifest for SpeechGuard AI — Samsung EnnovateX AX Hackathon 2026.
|
| 15 |
+
|
| 16 |
+
## Description
|
| 17 |
+
|
| 18 |
+
Evaluation protocol manifest for the SpeechGuard AI keyword spotting
|
| 19 |
+
and speaker verification pipeline.
|
| 20 |
+
|
| 21 |
+
## Contents
|
| 22 |
+
|
| 23 |
+
- ablation_table.txt: Full ablation study results
|
| 24 |
+
- demo_scenarios.txt: 6 demo scenario results
|
| 25 |
+
- evaluation_config.json: Evaluation configuration
|
| 26 |
+
|
| 27 |
+
## Results
|
| 28 |
+
|
| 29 |
+
| Metric | Value |
|
| 30 |
+
|---|---|
|
| 31 |
+
| TA Clean | 99.0% |
|
| 32 |
+
| TA Noisy | 98.5% |
|
| 33 |
+
| Demo scenarios | 6/6 correct |
|
| 34 |
+
| Latency | 1.2ms |
|
| 35 |
+
|
| 36 |
+
## License
|
| 37 |
+
|
| 38 |
+
CC BY 4.0 — Free to use with attribution.
|
| 39 |
+
|
| 40 |
+
## Citation
|
| 41 |
+
|
| 42 |
+
Samsung EnnovateX AX Hackathon 2026 — Problem #04
|
| 43 |
+
Team: Placecomm Prophets (IIT Kharagpur)
|