#!/usr/bin/env python3 # -*- coding: utf-8 -*- # # Copyright 2016-2099 Ailemon.net # # This file is part of ASRT Speech Recognition Tool. # # ASRT is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # ASRT is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with ASRT. If not, see . # ============================================================================ """ @author: nl8590687 ASRT语音识别基于gRPC协议的API服务器程序 """ import argparse import time from concurrent import futures import grpc from assets.asrt_pb2_grpc import AsrtGrpcServiceServicer, add_AsrtGrpcServiceServicer_to_server from assets.asrt_pb2 import SpeechResponse, TextResponse from speech_model import ModelSpeech from speech_model_zoo import SpeechModel251BN from speech_features import Spectrogram from language_model3 import ModelLanguage from utils.ops import decode_wav_bytes API_STATUS_CODE_OK = 200000 # OK API_STATUS_CODE_OK_PART = 206000 # 部分结果OK，用于stream API_STATUS_CODE_CLIENT_ERROR = 400000 API_STATUS_CODE_CLIENT_ERROR_FORMAT = 400001 # 请求数据格式错误 API_STATUS_CODE_CLIENT_ERROR_CONFIG = 400002 # 请求数据配置不支持 API_STATUS_CODE_SERVER_ERROR = 500000 API_STATUS_CODE_SERVER_ERROR_RUNNING = 500001 # 服务器运行中出错 parser = argparse.ArgumentParser(description='ASRT gRPC Protocol API Service') parser.add_argument('--listen', default='0.0.0.0', type=str, help='the network to listen') parser.add_argument('--port', default='20002', type=str, help='the port to listen') args = parser.parse_args() AUDIO_LENGTH = 1600 AUDIO_FEATURE_LENGTH = 200 CHANNELS = 1 # 默认输出的拼音的表示大小是1428，即1427个拼音+1个空白块 OUTPUT_SIZE = 1428 sm251bn = SpeechModel251BN( input_shape=(AUDIO_LENGTH, AUDIO_FEATURE_LENGTH, CHANNELS), output_size=OUTPUT_SIZE ) feat = Spectrogram() ms = ModelSpeech(sm251bn, feat, max_label_length=64) ms.load_model('save_models/' + sm251bn.get_model_name() + '.model.h5') ml = ModelLanguage('model_language') ml.load_model() _ONE_DAY_IN_SECONDS = 60 * 60 * 24 class ApiService(AsrtGrpcServiceServicer): ''' 继承AsrtGrpcServiceServicer,实现hello方法 ''' def __init__(self): pass def Speech(self, request, context): ''' 具体实现Speech的方法, 并按照pb的返回对象构造SpeechResponse返回 :param request: :param context: :return: ''' wav_data = request.wav_data wav_samples = decode_wav_bytes(samples_data=wav_data.samples, channels=wav_data.channels, byte_width=wav_data.byte_width) result = ms.recognize_speech(wav_samples, wav_data.sample_rate) print("语音识别声学模型结果:", result) return SpeechResponse(status_code=API_STATUS_CODE_OK, status_message='', result_data=result) def Language(self, request, context): ''' 具体实现Language的方法, 并按照pb的返回对象构造TextResponse返回 :param request: :param context: :return: ''' print('Language收到了请求:', request) result = ml.pinyin_to_text(list(request.pinyins)) print('Language结果:', result) return TextResponse(status_code=API_STATUS_CODE_OK, status_message='', text_result=result) def All(self, request, context): ''' 具体实现All的方法, 并按照pb的返回对象构造TextResponse返回 :param request: :param context: :return: ''' wav_data = request.wav_data wav_samples = decode_wav_bytes(samples_data=wav_data.samples, channels=wav_data.channels, byte_width=wav_data.byte_width) result_speech = ms.recognize_speech(wav_samples, wav_data.sample_rate) result = ml.pinyin_to_text(result_speech) print("语音识别结果:", result) return TextResponse(status_code=API_STATUS_CODE_OK, status_message='', text_result=result) def Stream(self, request_iterator, context): ''' 具体实现Stream的方法, 并按照pb的返回对象构造TextResponse返回 :param request: :param context: :return: ''' result = list() tmp_result_last = list() beam_size = 100 for request in request_iterator: wav_data = request.wav_data wav_samples = decode_wav_bytes(samples_data=wav_data.samples, channels=wav_data.channels, byte_width=wav_data.byte_width) result_speech = ms.recognize_speech(wav_samples, wav_data.sample_rate) for item_pinyin in result_speech: tmp_result = ml.pinyin_stream_decode(tmp_result_last, item_pinyin, beam_size) if len(tmp_result) == 0 and len(tmp_result_last) > 0: result.append(tmp_result_last[0][0]) print("流式语音识别结果：", ''.join(result)) yield TextResponse(status_code=API_STATUS_CODE_OK, status_message='', text_result=''.join(result)) result = list() tmp_result = ml.pinyin_stream_decode([], item_pinyin, beam_size) tmp_result_last = tmp_result yield TextResponse(status_code=API_STATUS_CODE_OK_PART, status_message='', text_result=''.join(tmp_result[0][0])) if len(tmp_result_last) > 0: result.append(tmp_result_last[0][0]) print("流式语音识别结果：", ''.join(result)) yield TextResponse(status_code=API_STATUS_CODE_OK, status_message='', text_result=''.join(result)) def run(host, port): ''' gRPC API服务启动 :return: ''' server = grpc.server(futures.ThreadPoolExecutor(max_workers=10)) add_AsrtGrpcServiceServicer_to_server(ApiService(),server) server.add_insecure_port(''.join([host, ':', port])) server.start() print("start service...") try: while True: time.sleep(_ONE_DAY_IN_SECONDS) except KeyboardInterrupt: server.stop(0) if __name__ == '__main__': run(host=args.listen, port=args.port)