LableniBot/utils.py at main · acw-upv/LableniBot · GitHub

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
from datetime import datetime
import numpy as np
import wave
import pyaudio


def get_current_time(only_unix=False, get_time_str=False):
    t = datetime.now()

    if only_unix:
        return t.timestamp()

    year = str(t.year)

    month = str(t.month)
    month = "0" + month if len(month) == 1 else month

    day = str(t.day)
    day = "0" + day if len(day) == 1 else day

    hour = str(t.hour)
    hour = "0" + hour if len(hour) == 1 else hour

    mint = str(t.minute)
    mint = "0" + mint if len(mint) == 1 else mint

    sec = str(t.second)
    sec = "0" + sec if len(sec) == 1 else sec

    mil_sec = str(t.microsecond)

    date_str = year + "-" + month + "-" + day + " " + hour + ":" + mint + ":" + sec + "." + mil_sec

    if get_time_str:
        return year + month + day + "_" + hour + mint + sec

    return date_str, t.timestamp()


def get_google_s2t_sent(example1_sent, example1_time):
    matrix_content = np.zeros((len(example1_sent), len(example1_sent)))

    for i, i_sent in enumerate(example1_sent):
        for j, j_sent in enumerate(example1_sent):
            matrix_content[i][j] = 1 if i_sent in j_sent else 0

    keep_vec_bool = matrix_content.sum(axis=0) > 1

    if np.sum(keep_vec_bool) > 0:
        order_idx = np.argsort(example1_time[keep_vec_bool])
        final_sentence = " ".join(example1_sent[keep_vec_bool][order_idx])
    else:
        final_sentence = " ".join(example1_sent)

    return final_sentence


def check_repetition_s2t(response_results):
    unique_sentences, unique_time = [], []
    for res in response_results:
        text_i = res.alternatives[0].transcript
        if text_i not in unique_sentences:
            unique_sentences.append(text_i)
            unique_time.append(res.result_end_time.total_seconds())

    return np.array(unique_sentences), np.array(unique_time)


def process_googles2t_answer(resp_res):
    # If 'response' has something inside.
    if len(resp_res) > 0:
        repeat_message_label = False

        if len(resp_res) > 1:

            uniq_sentences_array, uniq_time_array = check_repetition_s2t(resp_res)

            if len(uniq_sentences_array) > 1:
                # There is more than one answer in Google S2T result.
                spanish_text = get_google_s2t_sent(uniq_sentences_array, uniq_time_array)
            else:
                # There is only one answer in Google S2T result.
                spanish_text = uniq_sentences_array[0]

        else:
            spanish_text = resp_res[0].alternatives[0].transcript

    else:
        # Here means that Google S2T did not find any message.
        spanish_text = " "
        repeat_message_label = True

    return spanish_text, repeat_message_label


def reproduce_audio(root_bot_audio, chunk):
    # ###############
    # ### PYAUDIO ###
    # ###############

    # open a wav format music
    f = wave.open(root_bot_audio, "rb")
    # instantiate PyAudio
    p = pyaudio.PyAudio()
    # open stream
    stream = p.open(format=p.get_format_from_width(f.getsampwidth()),
                    channels=f.getnchannels(),
                    rate=f.getframerate(),
                    output=True)
    # read data
    data = f.readframes(chunk)

    # play stream
    while data:
        stream.write(data)
        data = f.readframes(chunk)
        # stop stream
    stream.stop_stream()
    stream.close()

    # close PyAudio
    p.terminate()