Whisper Large-v3 Turbo

Diese Soniqo-Seite dokumentiert Whisper Large-v3 Turbo aus der lokalen speech-swift- / speech-core-Implementierung. Hugging-Face-Bundles sind nach den Integrationshinweisen verlinkt.

Zuerst interne Seite

Landing-Karten und Docs-Menüs führen zuerst hierher; Quellen- und Bundle-Links bleiben auf dieser Seite verfügbar.

Überblick

Modell	Whisper Large-v3 Turbo
Rolle	General multilingual speech-to-text
Backend	CoreML fp16 on CPU and Neural Engine
Ausgabe	Text transcript
Sprachen	Whisper multilingual set, about 100 languages
Lizenz	MIT weights from OpenAI
Status	Ready through speech transcribe --engine whisper and the WhisperASR Swift product
Quelle	openai/whisper-large-v3-turbo
Swift-Produkt	`WhisperASR`
CLI / Laufzeit	`speech transcribe --engine whisper`

Verwendung

Das folgende Snippet entspricht der aktuellen API oder dem aktuellen Befehl aus speech-swift.

# Transcribe with the native CoreML Whisper runtime.
.build/release/speech transcribe recording.wav --engine whisper
.build/release/speech transcribe recording.wav --engine whisper --language en

Modelllinks

Implementierungsnotizen

LibriSpeech test-clean slice on an M5 Pro: 1.40% WER, mean RTF 0.089, 6.1 s model load, 384 MB peak RSS — versus 1.53% / 0.085 / 100.2 s / 507 MB for a direct WhisperKit run of the same model.
The bundle splits into four CoreML models — mel spectrogram, audio encoder, decoder context prefill, and a KV-cached text decoder — with the encoder and decoder on the Neural Engine.
Greedy no-timestamp decoding with automatic language detection or a --language hint; audio is processed in 30 s chunks with a repeated-word guard against greedy hallucination loops.
Word timestamps, temperature fallback, and VAD-guided long-form seeking are not implemented yet; --model accepts turbo, large-v3-turbo, or any Hugging Face repo ID.