File size: 7,488 Bytes
97ff100
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
from sys import platform
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
import logging
import torch
from transformers.utils import is_flash_attn_2_available
from pyannote.audio import Pipeline
from pyannote.core import Segment
import pandas as pd

languages = {
    "English": "en",
    "Chinese": "zh",
    "German": "de",
    "Spanish": "es",
    "Russian": "ru",
    "Korean": "ko",
    "French": "fr",
    "Japanese": "ja",
    "Portuguese": "pt",
    "Turkish": "tr",
    "Polish": "pl",
    "Catalan": "ca",
    "Dutch": "nl",
    "Arabic": "ar",
    "Swedish": "sv",
    "Italian": "it",
    "Indonesian": "id",
    "Hindi": "hi",
    "Finnish": "fi",
    "Vietnamese": "vi",
    "Hebrew": "iw",
    "Ukrainian": "uk",
    "Greek": "el",
    "Malay": "ms",
    "Czech": "cs",
    "Romanian": "ro",
    "Danish": "da",
    "Hungarian": "hu",
    "Tamil": "ta",
    "Norwegian": "no",
    "Thai": "th",
    "Urdu": "ur",
    "Croatian": "hr",
    "Bulgarian": "bg",
    "Lithuanian": "lt",
    "Latin": "la",
    "Maori": "mi",
    "Malayalam": "ml",
    "Welsh": "cy",
    "Slovak": "sk",
    "Telugu": "te",
    "Persian": "fa",
    "Latvian": "lv",
    "Bengali": "bn",
    "Serbian": "sr",
    "Azerbaijani": "az",
    "Slovenian": "sl",
    "Kannada": "kn",
    "Estonian": "et",
    "Macedonian": "mk",
    "Breton": "br",
    "Basque": "eu",
    "Icelandic": "is",
    "Armenian": "hy",
    "Nepali": "ne",
    "Mongolian": "mn",
    "Bosnian": "bs",
    "Kazakh": "kk",
    "Albanian": "sq",
    "Swahili": "sw",
    "Galician": "gl",
    "Marathi": "mr",
    "Punjabi": "pa",
    "Sinhala": "si",
    "Khmer": "km",
    "Shona": "sn",
    "Yoruba": "yo",
    "Somali": "so",
    "Afrikaans": "af",
    "Occitan": "oc",
    "Georgian": "ka",
    "Belarusian": "be",
    "Tajik": "tg",
    "Sindhi": "sd",
    "Gujarati": "gu",
    "Amharic": "am",
    "Yiddish": "yi",
    "Lao": "lo",
    "Uzbek": "uz",
    "Faroese": "fo",
    "Haitian creole": "ht",
    "Pashto": "ps",
    "Turkmen": "tk",
    "Nynorsk": "nn",
    "Maltese": "mt",
    "Sanskrit": "sa",
    "Luxembourgish": "lb",
    "Myanmar": "my",
    "Tibetan": "bo",
    "Tagalog": "tl",
    "Malagasy": "mg",
    "Assamese": "as",
    "Tatar": "tt",
    "Hawaiian": "haw",
    "Lingala": "ln",
    "Hausa": "ha",
    "Bashkir": "ba",
    "Javanese": "jw",
    "Sundanese": "su",
}

if torch.cuda.is_available():
    device = torch.device("cuda:0")
elif platform == "darwin":
    device = torch.device("mps")
else:
    device = torch.device("cpu")

torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32



def get_text_with_timestamp(transcribe_res):
    timestamp_texts = []
    for item in transcribe_res["chunks"]:
        start = item["timestamp"][0]
        end = item["timestamp"][1]
        text = item["text"]
        timestamp_texts.append((Segment(start, end), text))
    return timestamp_texts


def add_speaker_info_to_text(timestamp_texts, ann):
    spk_text = []
    for seg, text in timestamp_texts:
        spk = ann.crop(seg).argmax()
        spk_text.append((seg, spk, text))
    return spk_text


def merge_cache(text_cache):
    sentence = "".join([item[-1] for item in text_cache])
    spk = text_cache[0][1]
    start = text_cache[0][0].start
    end = text_cache[-1][0].end
    return Segment(start, end), spk, sentence


PUNC_SENT_END = [".", "?", "!"]


def merge_sentence(spk_text):
    merged_spk_text = []
    pre_spk = None
    text_cache = []
    for seg, spk, text in spk_text:
        if spk != pre_spk and pre_spk is not None and len(text_cache) > 0:
            merged_spk_text.append(merge_cache(text_cache))
            text_cache = [(seg, spk, text)]
            pre_spk = spk

        elif text[-1] in PUNC_SENT_END:
            text_cache.append((seg, spk, text))
            merged_spk_text.append(merge_cache(text_cache))
            text_cache = []
            pre_spk = spk
        else:
            text_cache.append((seg, spk, text))
            pre_spk = spk
    if len(text_cache) > 0:
        merged_spk_text.append(merge_cache(text_cache))
    return merged_spk_text

def diarize_text(transcribe_res, diarization_result):
    timestamp_texts = get_text_with_timestamp(transcribe_res)
    spk_text = add_speaker_info_to_text(timestamp_texts, diarization_result)
    res_processed = merge_sentence(spk_text)
    return res_processed

def make_conversation(transcribe_result, diarization_result):
    processed = diarize_text(transcribe_result, diarization_result)
    df = pd.DataFrame(processed, columns=["segment", "speaker", "text"])[
        ["speaker", "text"]
    ]
    df["key"] = (df["speaker"] != df["speaker"].shift(1)).astype(int).cumsum()
    conversation = df.groupby(["key", "speaker"])["text"].apply(" ".join).reset_index()
    conversation_list = list(zip(conversation.text, conversation.speaker))
    return conversation_list

# def transcriber(input: str, language: str, translate: bool, progress) -> dict:
def transcriber(input: str, model: str, language: str, translate: bool, diarize: bool, input_diarization_token) -> dict:
    """Transcribes the audio using the OpenAI Whisper model.
    Args:
        input: file path to the audio file in any format
        language: name of the language in which the audio is recorded
        translate: boolean indicator to enable immediate translation
    Returns: transcription and segment-timestamps.
    """
    model_id = model

    if diarize:

        pipeline_diarization = Pipeline.from_pretrained(
            "pyannote/speaker-diarization-3.1",
            use_auth_token=input_diarization_token)

        # send pipeline to GPU (when available)
        pipeline_diarization.to(device)

        # apply pretrained pipeline
        diarization = pipeline_diarization(input)

    # print the result
    # for turn, _, speaker in diarization.itertracks(yield_label=True):
    #     print(f"start={turn.start:.1f}s stop={turn.end:.1f}s speaker_{speaker}")

    model = AutoModelForSpeechSeq2Seq.from_pretrained(
        model_id,
        torch_dtype=torch_dtype, 
        low_cpu_mem_usage=True, 
        use_safetensors=True,
        use_flash_attention_2=True if is_flash_attn_2_available() else False
    )

    print(device)

    model.to(device)

    processor = AutoProcessor.from_pretrained(model_id)

    language = languages.get(language, None)
    task = None
    if translate:
        task = "translate"

    pipe = pipeline(
        "automatic-speech-recognition",
        model=model,
        tokenizer=processor.tokenizer,
        feature_extractor=processor.feature_extractor,
        max_new_tokens=128,
        chunk_length_s=15,
        batch_size=16,
        return_timestamps=True,
        torch_dtype=torch_dtype,
        device=device,
        generate_kwargs={"task": task}
    )
    

    results = pipe(input)
    results["text"] = results["text"].strip()
 
    text = ""
    chunks = results.get("chunks", [])
    for chunk in chunks:
        text += chunk["text"] + "\n"

    # conversation = make_conversation(transcription, diarization)

    # Transform the list to skip one line each time
    # conversation_gradio = []
    # for i in range(0, len(conversation), 2):  # Increment by 2 to skip one line each time
    #     current_text = conversation[i][0]
    #     next_text = conversation[i + 1][0] if i + 1 < len(conversation) else ""
    #     conversation_gradio.append((current_text, next_text))

    return text