Merge pull request #290 from veelion/main

Read audio_data to buf when speaking is False for non-stream inferring

funasr/runtime/grpc/Readme.md

@@ -53,6 +53,68 @@ cd ../python/grpc
 python grpc_main_client_mic.py  --host $server_ip --port 10108
 ```
 
+The `grpc_main_client_mic.py` follows the [original design] (https://github.com/alibaba-damo-academy/FunASR/tree/main/funasr/runtime/python/grpc#workflow-in-desgin) by sending audio_data with chunks. If you want to send audio_data in one request, here is an example:
+
+```
+# go to ../python/grpc to find this package
+import paraformer_pb2
+
+
+class RecognizeStub:
+    def __init__(self, channel):
+        self.Recognize = channel.stream_stream(
+                '/paraformer.ASR/Recognize',
+                request_serializer=paraformer_pb2.Request.SerializeToString,
+                response_deserializer=paraformer_pb2.Response.FromString,
+                )
+
+
+async def send(channel, data, speaking, isEnd):
+    stub = RecognizeStub(channel)
+    req = paraformer_pb2.Request()
+    if data:
+        req.audio_data = data
+    req.user = 'zz'
+    req.language = 'zh-CN'
+    req.speaking = speaking
+    req.isEnd = isEnd
+    q = queue.SimpleQueue()
+    q.put(req)
+    return stub.Recognize(iter(q.get, None))
+
+# send the audio data once
+async def grpc_rec(data, grpc_uri):
+    with grpc.insecure_channel(grpc_uri) as channel:
+        b = time.time()
+        response = await send(channel, data, False, False)
+        resp = response.next()
+        text = ''
+        if 'decoding' == resp.action:
+            resp = response.next()
+            if 'finish' == resp.action:
+                text = json.loads(resp.sentence)['text']
+        response = await send(channel, None, False, True)
+        return {
+                'text': text,
+                'time': time.time() - b,
+                }
+
+async def test():
+    # fc = FunAsrGrpcClient('127.0.0.1', 9900)
+    # t = await fc.rec(wav.tobytes())
+    # print(t)
+    wav, _ = sf.read('z-10s.wav', dtype='int16')
+    uri = '127.0.0.1:9900'
+    res = await grpc_rec(wav.tobytes(), uri)
+    print(res)
+
+
+if __name__ == '__main__':
+    asyncio.run(test())
+
+```
+
+
 ## Acknowledge
 1. This project is maintained by [FunASR community](https://github.com/alibaba-damo-academy/FunASR).
 2. We acknowledge [DeepScience](https://www.deepscience.cn) for contributing the grpc service.

funasr/runtime/grpc/paraformer_server.cc

@@ -88,7 +88,7 @@ grpc::Status ASRServicer::Recognize(
             res.set_language(req.language());
             stream->Write(res);
         } else if (!req.speaking()) {
-            if (client_buffers.count(req.user()) == 0) {
+            if (client_buffers.count(req.user()) == 0 && req.audio_data().size() == 0) {
                 Response res;
                 res.set_sentence(
                     R"({"success": true, "detail": "waiting_for_voice"})"
@@ -99,14 +99,18 @@ grpc::Status ASRServicer::Recognize(
                 stream->Write(res);
             }else {
                 auto begin_time = std::chrono::duration_cast<std::chrono::milliseconds>(std::chrono::system_clock::now().time_since_epoch()).count();
+                if (req.audio_data().size() > 0) {
+                  auto& buf = client_buffers[req.user()];
+                  buf.insert(buf.end(), req.audio_data().begin(), req.audio_data().end());
+                }
                 std::string tmp_data = this->client_buffers[req.user()];
                 this->clear_states(req.user());
-                
+
                 Response res;
                 res.set_sentence(
                     R"({"success": true, "detail": "decoding data: " + std::to_string(tmp_data.length()) + " bytes"})"
                 );
-		int data_len_int = tmp_data.length();
+                int data_len_int = tmp_data.length();
                 std::string data_len = std::to_string(data_len_int);
                 std::stringstream ss;
                 ss << R"({"success": true, "detail": "decoding data: )" << data_len << R"( bytes")"  << R"("})";
@@ -129,18 +133,18 @@ grpc::Status ASRServicer::Recognize(
                     res.set_user(req.user());
                     res.set_action("finish");
                     res.set_language(req.language());
-                    
+
-                    
+
-                    
+
                     stream->Write(res);
                 }
                 else {
-                    RPASR_RESULT Result= RapidAsrRecogPCMBuffer(AsrHanlde, tmp_data.c_str(), data_len_int, RASR_NONE, NULL);   
+                    RPASR_RESULT Result= RapidAsrRecogPCMBuffer(AsrHanlde, tmp_data.c_str(), data_len_int, RASR_NONE, NULL);
                     std::string asr_result = ((RPASR_RECOG_RESULT*)Result)->msg;
 
                     auto end_time = std::chrono::duration_cast<std::chrono::milliseconds>(std::chrono::system_clock::now().time_since_epoch()).count();
                     std::string delay_str = std::to_string(end_time - begin_time);
-                    
+
                     std::cout << "user: " << req.user() << " , delay(ms): " << delay_str << ", text: " << asr_result << std::endl;
                     Response res;
                     std::stringstream ss;
@@ -150,8 +154,8 @@ grpc::Status ASRServicer::Recognize(
                     res.set_user(req.user());
                     res.set_action("finish");
                     res.set_language(req.language());
-                    
+
-                    
+
                     stream->Write(res);
                 }
             }
@@ -165,7 +169,7 @@ grpc::Status ASRServicer::Recognize(
             res.set_language(req.language());
             stream->Write(res);
         }
-    }    
+    }
     return Status::OK;
 }