OpenAI 开源语音识别模型，99 种语言，准确率 95%+。
pip install openai-whisper命令行: whisper audio.mp3 --model large-v3 --language zhPython: import whisper
model = whisper.load_model('base')
result = model.transcribe('audio.mp3')from openai import OpenAI
client = OpenAI(api_key='...')audio_file = open('audio.mp3', 'rb')
transcript = client.audio.transcriptions.create(model='whisper-1', file=audio_file)价格: $0.006/分钟