### 实现 Python 中实时语音转文字并集成阿里云语音识别 API 为了实现实时语音转文字的功能，可以采用阿里云提供的 Paraformer 语音识别 API。此方案不仅支持高精度的录音转写，还具备良好的扩展性和易用性。 #### 准备工作 确保已注册阿里云账户，并获取访问密钥（AccessKey ID 和 AccessKey Secret），这些信息用于身份验证和权限管理[^2]。 #### 安装依赖库 安装必要的 Python 库来简化与阿里云的服务交互： ```bash pip install aliyun-python-sdk-core pip install pyaudio pip install requests ``` #### 编写代码实现 下面展示了一个简单的例子，说明如何利用麦克风输入流并通过阿里云 API 将其转换成文本形式： ```python import json from aliyunsdkcore.client import AcsClient from aliyunsdkcore.request import CommonRequest import pyaudio import wave def get_token(): client = AcsClient('<your-access-key-id>', '<your-access-key-secret>', 'cn-hangzhou') request = CommonRequest() request.set_accept_format('json') request.set_domain('nls-meta.cn-hangzhou.aliyuncs.com') request.set_method('POST') request.set_version('2019-02-28') request.set_action_name('CreateToken') response = client.do_action_with_exception(request) token_info = json.loads(response.decode()) return token_info['Token']['Id'] def recognize_speech(token_id, audio_file_path): url = "https://nls-gateway.cn-shanghai.aliyuncs.com/stream/v1/asr" headers = { 'X-NLS-Token': token_id, 'Content-Type': 'application/octet-stream' } with open(audio_file_path, 'rb') as f: data = f.read() params = ( ('appkey', '<your-app-key>'), ('format', 'pcm'), ('sample_rate', 16000), ('enable_intermediate_result', True), ('enable_punctuation_prediction', True), ('enable_inverse_text_normalization', True) ) resp = requests.post(url, headers=headers, params=params, data=data).text result_json = json.loads(resp) if 'result' in result_json and isinstance(result_json['result'], str): print(f"Recognized Text: {result_json['result']}") return result_json['result'] else: raise Exception("Failed to Recognize Speech") if __name__ == "__main__": CHUNK = 1024 FORMAT = pyaudio.paInt16 CHANNELS = 1 RATE = 16000 RECORD_SECONDS = 5 p = pyaudio.PyAudio() stream = p.open(format=FORMAT, channels=CHANNELS, rate=RATE, input=True, frames_per_buffer=CHUNK) print("* recording") frames = [] for i in range(0, int(RATE / CHUNK * RECORD_SECONDS)): data = stream.read(CHUNK) frames.append(data) print("* done recording") stream.stop_stream() stream.close() p.terminate() wf = wave.open("output.wav", 'wb') wf.setnchannels(CHANNELS) wf.setsampwidth(p.get_sample_size(FORMAT)) wf.setframerate(RATE) wf.writeframes(b''.join(frames)) wf.close() token = get_token() recognized_text = recognize_speech(token, "./output.wav") ``` 上述脚本首先录制一段来自默认音频设备的声音片段保存为 WAV 文件，接着调用 `recognize_speech` 方法上传该文件至阿里云服务器完成语音识别过程。