|
|
--- |
|
|
license: cc-by-nc-4.0 |
|
|
task_categories: |
|
|
- audio-classification |
|
|
language: |
|
|
- en |
|
|
tags: |
|
|
- speaker-verification |
|
|
- speaker-recognition |
|
|
- audio |
|
|
- benchmark |
|
|
pretty_name: CASE Benchmark |
|
|
size_categories: |
|
|
- 10K<n<100K |
|
|
--- |
|
|
|
|
|
# CASE Benchmark |
|
|
|
|
|
**Carrier-Agnostic Speaker Embedding Benchmark** - A comprehensive benchmark for evaluating speaker verification systems under real-world acoustic carrier conditions. |
|
|
|
|
|
## Overview |
|
|
|
|
|
CASE Benchmark tests speaker embedding models across 24 protocols covering: |
|
|
- **Clean**: Matched enrollment and test conditions |
|
|
- **Codecs** (7): GSM, G.711 μ-law/A-law, Opus (6k/12k/24k), MP3 |
|
|
- **Microphones** (7): Laptop, webcam, phone, headset, conference |
|
|
- **Noise** (5): SNR levels from 5dB to 25dB |
|
|
- **Reverb** (1): Simulated room acoustics |
|
|
- **Playback chains** (3): Combined codec + noise + microphone |
|
|
|
|
|
## Quick Start |
|
|
|
|
|
```python |
|
|
from case_benchmark.download import download_benchmark |
|
|
|
|
|
# Download full benchmark (~3.1GB compressed) |
|
|
download_benchmark("./benchmark") |
|
|
|
|
|
# Download only specific conditions |
|
|
download_benchmark("./benchmark", conditions=["clean", "codec"]) |
|
|
|
|
|
# Download only VoxCeleb1-O dataset |
|
|
download_benchmark("./benchmark", datasets=["voxceleb1_o"]) |
|
|
``` |
|
|
|
|
|
## Structure |
|
|
|
|
|
Audio files are stored as compressed archives per condition: |
|
|
|
|
|
``` |
|
|
audio/ |
|
|
├── voxceleb1_o_clean.tar.gz # 69MB - 400 files |
|
|
├── voxceleb1_o_codec.tar.gz # 464MB - 2,800 files |
|
|
├── voxceleb1_o_mic.tar.gz # 467MB - 2,800 files |
|
|
├── voxceleb1_o_noise.tar.gz # 346MB - 2,000 files |
|
|
├── voxceleb1_o_reverb.tar.gz # 71MB - 400 files |
|
|
├── voxceleb1_o_playback.tar.gz # 205MB - 1,200 files |
|
|
├── librispeech_clean.tar.gz # 63MB - 392 files |
|
|
├── librispeech_codec.tar.gz # 426MB - 2,744 files |
|
|
├── librispeech_mic.tar.gz # 430MB - 2,744 files |
|
|
├── librispeech_noise.tar.gz # 331MB - 1,960 files |
|
|
├── librispeech_reverb.tar.gz # 67MB - 392 files |
|
|
└── librispeech_playback.tar.gz # 196MB - 1,176 files |
|
|
|
|
|
trials/ |
|
|
├── clean_clean.txt |
|
|
├── clean_codec_*.txt # 7 codec protocols |
|
|
├── clean_mic_*.txt # 7 microphone protocols |
|
|
├── clean_noise_*.txt # 5 noise protocols |
|
|
├── clean_reverb.txt |
|
|
└── clean_playback_*.txt # 3 playback chain protocols |
|
|
``` |
|
|
|
|
|
After extraction, audio files are organized as: |
|
|
``` |
|
|
voxceleb1_o/ |
|
|
├── clean/{speaker_id}/utt_*.wav |
|
|
├── codec/{codec_type}/{speaker_id}/utt_*.wav |
|
|
├── mic/{mic_type}/{speaker_id}/utt_*.wav |
|
|
├── noise/{snr_level}/{speaker_id}/utt_*.wav |
|
|
├── reverb/{speaker_id}/utt_*.wav |
|
|
└── playback/{chain_type}/{speaker_id}/utt_*.wav |
|
|
``` |
|
|
|
|
|
## Trial Format |
|
|
|
|
|
Each trial file contains lines in the format: |
|
|
``` |
|
|
<label> <enrollment_path> <test_path> |
|
|
``` |
|
|
|
|
|
Where: |
|
|
- `label`: 1 for same speaker, 0 for different speaker |
|
|
- `enrollment_path`: Path to enrollment audio (always clean) |
|
|
- `test_path`: Path to test audio (condition-dependent) |
|
|
|
|
|
## Datasets |
|
|
|
|
|
| Dataset | Speakers | Utterances | Source | |
|
|
|---------|----------|------------|--------| |
|
|
| VoxCeleb1-O | 40 | 400 clean | VoxCeleb1 test set | |
|
|
| LibriSpeech | 40 | 392 clean | LibriSpeech test-clean | |
|
|
|
|
|
## Leaderboard |
|
|
|
|
|
| Rank | Model | Absolute EER | Degradation | Clean EER | |
|
|
|------|-------|-------------|-------------|-----------| |
|
|
| 1 | WeSpeaker ResNet34 | **3.01%** | +2.43% | 0.58% | |
|
|
| 2 | SpeechBrain ECAPA-TDNN | 3.05% | +2.49% | 0.56% | |
|
|
| 3 | CASE HF v2-512 | 3.53% | **+2.31%** | 1.22% | |
|
|
| 4 | NeMo TitaNet-L | 4.05% | +3.39% | 0.66% | |
|
|
| 5 | pyannote Embedding | 4.47% | +2.79% | 1.68% | |
|
|
| 6 | Resemblyzer | 10.49% | +5.65% | 4.84% | |
|
|
|
|
|
See [full results](https://github.com/gittb/case-benchmark/tree/master/results) for detailed per-protocol breakdowns. |
|
|
|
|
|
## Related Resources |
|
|
|
|
|
| Resource | Description | Link | |
|
|
|----------|-------------|------| |
|
|
| **CASE HF v2-512** | Carrier-agnostic speaker embedding model | [HuggingFace Model](https://huggingface.co/gittb/case-speaker-embedding-v2) | |
|
|
| **Benchmark Code** | Evaluation scripts and tools | [GitHub](https://github.com/gittb/case-benchmark) | |
|
|
| **Metrics Guide** | How to interpret Clean EER, Degradation Factor | [Metrics Documentation](https://github.com/gittb/case-benchmark/blob/master/docs/metrics.md) | |
|
|
| **Submission Guide** | How to submit your model to the leaderboard | [Submission Guide](https://github.com/gittb/case-benchmark/blob/master/docs/submission.md) | |
|
|
|
|
|
## License |
|
|
|
|
|
- **Audio data**: CC BY-NC 4.0 (non-commercial use) |
|
|
- **Evaluation code**: MIT License |
|
|
|
|
|
## Citation |
|
|
|
|
|
```bibtex |
|
|
@misc{case-benchmark-2026, |
|
|
title={CASE Benchmark: Carrier-Agnostic Speaker Embedding Evaluation}, |
|
|
author={Gitter, Ben}, |
|
|
year={2026}, |
|
|
url={https://github.com/gittb/case-benchmark} |
|
|
} |
|
|
``` |
|
|
|
