Assembling...
Distilled Whisper variant that retains large-v3 accuracy while running significantly faster. Reduces decoder layers for efficient inference on consumer GPUs and edge devices while preserving multilingual robustness.
Transcript will appear here in real-time as you speak…
Near large-v3 accuracy at a fraction of the compute cost
Runs real-time on consumer GPUs and Apple Silicon
MIT license with full Whisper ecosystem compatibility
Transcribe meetings in real-time with speaker identification and punctuation.
Analyze customer calls at scale with sentiment detection and keyword spotting.
Convert audio and video libraries into searchable text archives.
Provide real-time captions for broadcasts, presentations, and live events.
// Whisper large-v3-turbo — Speech-to-Text
import { transcribe } from "@arkitekton/voice";
const result = await transcribe({
model: "vm-oai-006",
vendor: "openai",
audio: audioFile,
language: "en",
options: {
punctuate: true,
diarize: true,
smart_format: true,
},
});
console.log("Transcript:", result.text);
console.log("Confidence:", result.confidence);Foundation models for real-time voice and transcription