Views
No views yet
1from optimum.intel import OVModelForSpeechSeq2Seq
2from transformers import AutoProcessor
3
4model_id = "CelesteImperia/Whisper-Large-v3-Turbo-OpenVINO-INT8"
5processor = AutoProcessor.from_pretrained(model_id)
6model = OVModelForSpeechSeq2Seq.from_pretrained(model_id)
7
8# Process local audio file
9audio_input = "path/to/audio.mp3"
10inputs = processor(audio_input, sampling_rate=16000, return_tensors="pt")
11generated_ids = model.generate(inputs.input_features, max_new_tokens=128)
12print(processor.batch_decode(generated_ids, skip_special_tokens=True)[0])WhisperPipeline in the OpenVINO.GenAI framework, enabling high-performance transcription in Windows applications.1using OpenVino.GenAI;
2
3// 1. Initialize the Transcription Pipeline
4var device = "CPU"; // Use "GPU" to leverage your hardware
5using var pipe = new WhisperPipeline("path/to/whisper-int8-model", device);
6
7// 2. Execute Transcription
8var result = pipe.Generate("path/to/audio.wav");
9Console.WriteLine(result.Texts[0]);| Platform | Support Link |
|---|---|
| Global & India | Support via Razorpay |
