Audio Transcribe

🚀 音頻轉錄模型運行指南

本指南將詳細介紹如何安裝必要的工具和依賴項，並運行音頻轉錄模型。

🚀 快速開始

安裝Chocolatey

在命令提示符（CMD）中運行以下命令來安裝Chocolatey：

@"%SystemRoot%\System32\WindowsPowerShell\v1.0\powershell.exe" -NoProfile -InputFormat None -ExecutionPolicy Bypass -Command "[System.Net.ServicePointManager]::SecurityProtocol = 3072; iex ((New-Object System.Net.WebClient).DownloadString('https://community.chocolatey.org/install.ps1'))" && SET "PATH=%PATH%;%ALLUSERSPROFILE%\chocolatey\bin"

安裝FFmpeg

安裝Chocolatey後，在CMD中運行以下命令來安裝FFmpeg：

choco install ffmpeg

安裝Python依賴項

在Python集成開發環境（IDE）中運行以下命令來安裝所需的依賴項：

pip install --upgrade pip
pip install --upgrade git+https://github.com/huggingface/transformers.git accelerate datasets[audio]

模型推理

運行以下代碼來進行模型推理：

import torch
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline

device = "cuda:0" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32

model_id = "washeed/audio-transcribe"

model = AutoModelForSpeechSeq2Seq.from_pretrained(
    model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
)
model.to(device)

processor = AutoProcessor.from_pretrained(model_id)

pipe = pipeline(
    "automatic-speech-recognition",
    model=model,
    tokenizer=processor.tokenizer,
    feature_extractor=processor.feature_extractor,
    max_new_tokens=128,
    chunk_length_s=30,
    batch_size=16,
    return_timestamps=True,
    torch_dtype=torch_dtype,
    device=device,
)

result = pipe("audio.mp3")
print(result["text"])

轉錄而非翻譯

如果您想進行轉錄而不是翻譯，只需將以下代碼：

result = pipe("audio.mp3")

替換為：

result = pipe("inference.mp3", generate_kwargs={"task": "transcribe"})