offline-websocket-client-decode-files-sequential.py
4.0 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
#!/usr/bin/env python3
#
# Copyright (c) 2023 Xiaomi Corporation
"""
A websocket client for sherpa-onnx-offline-websocket-server
This file shows how to use a single connection to transcribe multiple
files sequentially.
Usage:
./offline-websocket-client-decode-files-sequential.py \
--server-addr localhost \
--server-port 6006 \
/path/to/foo.wav \
/path/to/bar.wav \
/path/to/16kHz.wav \
/path/to/8kHz.wav
(Note: You have to first start the server before starting the client)
You can find the server at
https://github.com/k2-fsa/sherpa-onnx/blob/master/sherpa-onnx/csrc/offline-websocket-server.cc
Note: The server is implemented in C++.
"""
import argparse
import asyncio
import logging
import wave
from typing import List, Tuple
try:
import websockets
except ImportError:
print("please run:")
print("")
print(" pip install websockets")
print("")
print("before you run this script")
print("")
import numpy as np
def get_args():
parser = argparse.ArgumentParser(
formatter_class=argparse.ArgumentDefaultsHelpFormatter
)
parser.add_argument(
"--server-addr",
type=str,
default="localhost",
help="Address of the server",
)
parser.add_argument(
"--server-port",
type=int,
default=6006,
help="Port of the server",
)
parser.add_argument(
"sound_files",
type=str,
nargs="+",
help="The input sound file(s) to decode. Each file must be of WAVE"
"format with a single channel, and each sample has 16-bit, "
"i.e., int16_t. "
"The sample rate of the file can be arbitrary and does not need to "
"be 16 kHz",
)
return parser.parse_args()
def read_wave(wave_filename: str) -> Tuple[np.ndarray, int]:
"""
Args:
wave_filename:
Path to a wave file. It should be single channel and each sample should
be 16-bit. Its sample rate does not need to be 16kHz.
Returns:
Return a tuple containing:
- A 1-D array of dtype np.float32 containing the samples, which are
normalized to the range [-1, 1].
- sample rate of the wave file
"""
with wave.open(wave_filename) as f:
assert f.getnchannels() == 1, f.getnchannels()
assert f.getsampwidth() == 2, f.getsampwidth() # it is in bytes
num_samples = f.getnframes()
samples = f.readframes(num_samples)
samples_int16 = np.frombuffer(samples, dtype=np.int16)
samples_float32 = samples_int16.astype(np.float32)
samples_float32 = samples_float32 / 32768
return samples_float32, f.getframerate()
async def run(
server_addr: str,
server_port: int,
sound_files: List[str],
):
async with websockets.connect(
f"ws://{server_addr}:{server_port}"
) as websocket: # noqa
for wave_filename in sound_files:
logging.info(f"Sending {wave_filename}")
samples, sample_rate = read_wave(wave_filename)
assert isinstance(sample_rate, int)
assert samples.dtype == np.float32, samples.dtype
assert samples.ndim == 1, samples.dim
buf = sample_rate.to_bytes(4, byteorder="little") # 4 bytes
buf += (samples.size * 4).to_bytes(4, byteorder="little")
buf += samples.tobytes()
await websocket.send(buf)
decoding_results = await websocket.recv()
print(decoding_results)
# to signal that the client has sent all the data
await websocket.send("Done")
async def main():
args = get_args()
logging.info(vars(args))
server_addr = args.server_addr
server_port = args.server_port
sound_files = args.sound_files
await run(
server_addr=server_addr,
server_port=server_port,
sound_files=sound_files,
)
if __name__ == "__main__":
formatter = (
"%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s" # noqa
)
logging.basicConfig(format=formatter, level=logging.INFO)
asyncio.run(main())