Merge pull request #468 from alibaba-damo-academy/dev_websocket_offline

zhifu gao

2023-05-08 7557fa87a8043867e39f4920b38d6163d54095ef

Merge pull request #468 from alibaba-damo-academy/dev_websocket_offline

Dev websocket offline

 funasr/runtime/python/websocket/README.md

@@ -58,7 +58,7 @@
##### Loadding from wav.scp(kaldi style)
```shell
# --chunk_interval, "10": 600/10=60ms, "5"=600/5=120ms, "20": 600/12=30ms
python ws_client.py --host "0.0.0.0" --port 10095 --chunk_interval 10 --words_max_print 100 --audio_in "./data/wav.scp" --send_without_sleep
python ws_client.py --host "0.0.0.0" --port 10095 --chunk_interval 10 --words_max_print 100 --audio_in "./data/wav.scp" --send_without_sleep --output_dir "./results"
```
#### ASR streaming client
##### Recording from mircrophone
@@ -69,7 +69,7 @@
##### Loadding from wav.scp(kaldi style)
```shell
# --chunk_size, "5,10,5"=600ms, "8,8,4"=480ms
python ws_client.py --host "0.0.0.0" --port 10095 --chunk_size "5,10,5" --audio_in "./data/wav.scp" --words_max_print 100
python ws_client.py --host "0.0.0.0" --port 10095 --chunk_size "5,10,5" --audio_in "./data/wav.scp" --words_max_print 100 --output_dir "./results"
```

#### ASR offline/online 2pass client

 funasr/runtime/python/websocket/ws_client.py

@@ -7,7 +7,9 @@
import argparse
import json
import traceback
from multiprocessing import  Process
from multiprocessing import Process
from funasr.fileio.datadir_writer import DatadirWriter

parser = argparse.ArgumentParser()
parser.add_argument("--host",
                    type=str,
@@ -43,6 +45,10 @@
                    type=int,
                    default=100,
                    help="chunk")
parser.add_argument("--output_dir",
                    type=str,
                    default=None,
                    help="output_dir")

args = parser.parse_args()
args.chunk_size = [int(x) for x in args.chunk_size.split(",")]
@@ -50,6 +56,11 @@
# voices = asyncio.Queue()
from queue import Queue
voices = Queue()

ibest_writer = None
if args.output_dir is not None:
    writer = DatadirWriter(args.output_dir)
    ibest_writer = writer[f"1best_recog"]

async def record_microphone():
    is_finished = False
@@ -91,7 +102,9 @@
        wavs = [args.audio_in]
    for wav in wavs:
        wav_splits = wav.strip().split()
        wav_name = wav_splits[0] if len(wav_splits) > 1 else "demo"
        wav_path = wav_splits[1] if len(wav_splits) > 1 else wav_splits[0]
        
        # bytes_f = open(wav_path, "rb")
        # bytes_data = bytes_f.read()
        with wave.open(wav_path, "rb") as wav_file:
@@ -112,7 +125,7 @@
            beg = i*stride
            data = audio_bytes[beg:beg+stride]
            data = data.decode('ISO-8859-1')
            message = json.dumps({"chunk_size": args.chunk_size, "chunk_interval": args.chunk_interval, "is_speaking": is_speaking, "audio": data, "is_finished": is_finished})
            message = json.dumps({"chunk_size": args.chunk_size, "chunk_interval": args.chunk_interval, "is_speaking": is_speaking, "audio": data, "is_finished": is_finished, "wav_name": wav_name})
            voices.put(message)
            # print("data_chunk: ", len(data_chunk))
            # print(voices.qsize())
@@ -152,14 +165,18 @@
            # print(meg, end = '')
            # print("\r")
            # print(meg)
            wav_name = meg.get("wav_name", "demo")
            print(wav_name)
            text = meg["text"][0]
            if ibest_writer is not None:
                ibest_writer["text"][wav_name] = text
            if meg["mode"] == "online":
                text_print += " {}".format(text)
            else:
                text_print += "{}".format(text)
            text_print = text_print[-args.words_max_print:]
            os.system('clear')
            print("\r"+str(id)+":"+text_print)
            print("\rpid"+str(id)+": "+text_print)
        except Exception as e:
            print("Exception:", e)
            traceback.print_exc()

 funasr/runtime/python/websocket/ws_server_offline.py

@@ -78,6 +78,7 @@

                is_speaking = message["is_speaking"]
                websocket.param_dict_vad["is_final"] = not is_speaking
                websocket.wav_name = message.get("wav_name", "demo")
                if speech_start:
                    frames_asr.append(audio)
                speech_start_i, speech_end_i = await async_vad(websocket, audio)
@@ -136,8 +137,9 @@
                    rec_result = inference_pipeline_punc(text_in=rec_result['text'],
                                                         param_dict=websocket.param_dict_punc)
                    # print(rec_result)
                    message = json.dumps({"mode": "offline", "text": [rec_result["text"]]})
                    await websocket.send(message)
                message = json.dumps({"mode": "offline", "text": [rec_result["text"]], "wav_name": websocket.wav_name})
                await websocket.send(message)
                
                    
 


 funasr/runtime/python/websocket/ws_server_online.py

@@ -53,7 +53,7 @@

                is_speaking = message["is_speaking"]
                websocket.param_dict_asr_online["is_final"] = not is_speaking

                websocket.wav_name = message.get("wav_name", "demo")
                websocket.param_dict_asr_online["chunk_size"] = message["chunk_size"]
                
                frames_online.append(audio)
@@ -81,9 +81,9 @@
                    websocket.param_dict_asr_online["cache"] = dict()
                if "text" in rec_result:
                    if rec_result["text"] != "sil" and rec_result["text"] != "waiting_for_more_voice":
                        if len(rec_result["text"])>0:
                            rec_result["text"][0]=rec_result["text"][0] #.replace(" ","")
                        message = json.dumps({"mode": "online", "text": rec_result["text"]})
                        # if len(rec_result["text"])>0:
                        #     rec_result["text"][0]=rec_result["text"][0] #.replace(" ","")
                        message = json.dumps({"mode": "online", "text": rec_result["text"], "wav_name": websocket.wav_name})
                        await websocket.send(message)



 funasr/version.txt

@@ -1 +1 @@
0.4.5
0.4.6