Make result structs sendable for ASR

[BrandonWeng]

No description provided.

[github-actions]

Speaker Diarization Benchmark Results

Speaker Diarization Performance

Evaluating "who spoke when" detection accuracy

Metric	Value	Target	Status	Description
DER	18.7%	<30%	✅	Diarization Error Rate (lower is better)
JER	22.6%	<25%	✅	Jaccard Error Rate
RTF	0.05x	<1.0x	✅	Real-Time Factor (lower is faster)

Diarization Pipeline Timing Breakdown

Time spent in each stage of speaker diarization

Stage	Time (s)	%	Description
Model Download	0.000	0.0	Fetching diarization models
Model Compile	3.929	7.0	CoreML compilation
Audio Load	0.078	0.1	Loading audio file
Segmentation	13.615	24.2	Detecting speech regions
Embedding	38.572	68.6	Extracting speaker voices
Clustering	0.026	0.0	Grouping same speakers
Total	56.220	100	Full pipeline

Speaker Diarization Research Comparison

Research baselines typically achieve 18-30% DER on standard datasets

Method	DER	Notes
FluidAudio	18.7%	On-device CoreML
Research baseline	18-30%	Standard dataset performance

_{🎯 Speaker Diarization Test • AMI Corpus ES2004a • 1049.4s meeting audio • 52.2s diarization time • Test runtime: 1m 21s • 07/29/2025, 07:22 PM EST}

[github-actions]

VAD Benchmark Results

Performance Comparison

Dataset	Accuracy	Precision	Recall	F1-Score	RTFx	Files
MUSAN	100.0%	100.0%	100.0%	100.0%	13.4x faster	50
VOiCES	100.0%	100.0%	100.0%	100.0%	15.1x faster	50

Dataset Details

• MUSAN: Music, Speech, and Noise dataset - standard VAD evaluation
• VOiCES: Voices Obscured in Complex Environmental Settings - tests robustness in real-world conditions

✅ EXCELLENT: Average F1-Score above 70%

[github-actions]

ASR Benchmark Results

Dataset	WER Avg	WER Med	RTFx	Status
test-clean	4.07%	0.00%	1.50x	✅
test-other	8.26%	2.78%	1.44x	✅

_{500 files per dataset • Test runtime: 42m36s • 07/29/2025, 08:04 PM EST}

_{RTFx = Real-Time Factor (higher is better) • Calculated as: Total audio duration ÷ Total processing time
Processing time includes: Model inference on Apple Neural Engine, audio preprocessing, state resets between files, token-to-text conversion, and file I/O
Example: RTFx of 2.0x means 10 seconds of audio processed in 5 seconds (2x faster than real-time)}

Expected RTFx Performance on Physical M1 Hardware:

• M1 Mac: ~28x (clean), ~25x (other)
• CI shows ~0.5-3x due to virtualization limitations

_{Testing methodology follows HuggingFace Open ASR Leaderboard}