Apple's new SpeechAnalyzer beats Whisper Small on accuracy in first public benchmark

Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor

Inscribe benchmarked Apple's new SpeechAnalyzer API against the legacy SFSpeechRecognizer and three Whisper models on 5,559 LibriSpeech utterances. SpeechAnalyzer hit 2.12% WER on clean speech and 4.56% on noisy speech, beating Whisper Small by 1.62 and 3.39 points respectively while running ~3x faster. The legacy API scored 9.02% WER, worse than the 40MB Whisper Tiny. All engines ran fully on-device on an M2 Pro. Inscribe switched its default engine to SpeechAnalyzer and released all transcripts and scoring code. The post does not disclose SpeechAnalyzer's model architecture or parameter count.

Why it matters: First independent benchmark of Apple's SpeechAnalyzer with solid methodology (5,559 utterances, all on-device). Directly useful for voice product teams. Not 85+ because it's a single third-party benchmark on one dataset, not an Apple launch, and LibriSpeech alone doesn't cover...