Spaces:
Sleeping
Sleeping
Add prompt window to VAD
Browse filesThis allows each speech section to access the text of the previous
speech section if it ends within a certain time window.
- app.py +13 -10
- cli.py +3 -1
- src/utils.py +3 -3
- src/vad.py +106 -15
app.py
CHANGED
@@ -53,7 +53,7 @@ class WhisperTranscriber:
|
|
53 |
self.inputAudioMaxDuration = inputAudioMaxDuration
|
54 |
self.deleteUploadedFiles = deleteUploadedFiles
|
55 |
|
56 |
-
def transcribe_webui(self, modelName, languageName, urlData, uploadFile, microphoneData, task, vad, vadMergeWindow, vadMaxMergeSize, vadPadding):
|
57 |
try:
|
58 |
source, sourceName = self.__get_source(urlData, uploadFile, microphoneData)
|
59 |
|
@@ -68,7 +68,7 @@ class WhisperTranscriber:
|
|
68 |
self.model_cache[selectedModel] = model
|
69 |
|
70 |
# Execute whisper
|
71 |
-
result = self.transcribe_file(model, source, selectedLanguage, task, vad, vadMergeWindow, vadMaxMergeSize, vadPadding)
|
72 |
|
73 |
# Write result
|
74 |
downloadDirectory = tempfile.mkdtemp()
|
@@ -88,9 +88,9 @@ class WhisperTranscriber:
|
|
88 |
return [], ("[ERROR]: Maximum remote video length is " + str(e.maxDuration) + "s, file was " + str(e.videoDuration) + "s"), "[ERROR]"
|
89 |
|
90 |
def transcribe_file(self, model: whisper.Whisper, audio_path: str, language: str, task: str = None, vad: str = None,
|
91 |
-
vadMergeWindow: float = 5, vadMaxMergeSize: float = 150, vadPadding: float = 1, **decodeOptions: dict):
|
92 |
# Callable for processing an audio file
|
93 |
-
whisperCallable = lambda audio : model.transcribe(audio, language=language, task=task, **decodeOptions)
|
94 |
|
95 |
# The results
|
96 |
if (vad == 'silero-vad'):
|
@@ -100,7 +100,8 @@ class WhisperTranscriber:
|
|
100 |
|
101 |
process_gaps = VadSileroTranscription(transcribe_non_speech = True,
|
102 |
max_silent_period=vadMergeWindow, max_merge_size=vadMaxMergeSize,
|
103 |
-
segment_padding_left=vadPadding, segment_padding_right=vadPadding,
|
|
|
104 |
result = process_gaps.transcribe(audio_path, whisperCallable)
|
105 |
elif (vad == 'silero-vad-skip-gaps'):
|
106 |
# Use Silero VAD
|
@@ -109,7 +110,8 @@ class WhisperTranscriber:
|
|
109 |
|
110 |
skip_gaps = VadSileroTranscription(transcribe_non_speech = False,
|
111 |
max_silent_period=vadMergeWindow, max_merge_size=vadMaxMergeSize,
|
112 |
-
segment_padding_left=vadPadding, segment_padding_right=vadPadding,
|
|
|
113 |
result = skip_gaps.transcribe(audio_path, whisperCallable)
|
114 |
elif (vad == 'periodic-vad'):
|
115 |
# Very simple VAD - mark every 5 minutes as speech. This makes it less likely that Whisper enters an infinite loop, but
|
@@ -118,7 +120,7 @@ class WhisperTranscriber:
|
|
118 |
result = periodic_vad.transcribe(audio_path, whisperCallable)
|
119 |
else:
|
120 |
# Default VAD
|
121 |
-
result = whisperCallable(audio_path)
|
122 |
|
123 |
return result
|
124 |
|
@@ -217,9 +219,10 @@ def create_ui(inputAudioMaxDuration, share=False, server_name: str = None):
|
|
217 |
gr.Audio(source="microphone", type="filepath", label="Microphone Input"),
|
218 |
gr.Dropdown(choices=["transcribe", "translate"], label="Task"),
|
219 |
gr.Dropdown(choices=["none", "silero-vad", "silero-vad-skip-gaps", "periodic-vad"], label="VAD"),
|
220 |
-
gr.Number(label="VAD - Merge Window (s)", precision=0, value=
|
221 |
-
gr.Number(label="VAD - Max Merge Size (s)", precision=0, value=
|
222 |
-
gr.Number(label="VAD - Padding (s)", precision=None, value=1)
|
|
|
223 |
], outputs=[
|
224 |
gr.File(label="Download"),
|
225 |
gr.Text(label="Transcription"),
|
|
|
53 |
self.inputAudioMaxDuration = inputAudioMaxDuration
|
54 |
self.deleteUploadedFiles = deleteUploadedFiles
|
55 |
|
56 |
+
def transcribe_webui(self, modelName, languageName, urlData, uploadFile, microphoneData, task, vad, vadMergeWindow, vadMaxMergeSize, vadPadding, vadPromptWindow):
|
57 |
try:
|
58 |
source, sourceName = self.__get_source(urlData, uploadFile, microphoneData)
|
59 |
|
|
|
68 |
self.model_cache[selectedModel] = model
|
69 |
|
70 |
# Execute whisper
|
71 |
+
result = self.transcribe_file(model, source, selectedLanguage, task, vad, vadMergeWindow, vadMaxMergeSize, vadPadding, vadPromptWindow)
|
72 |
|
73 |
# Write result
|
74 |
downloadDirectory = tempfile.mkdtemp()
|
|
|
88 |
return [], ("[ERROR]: Maximum remote video length is " + str(e.maxDuration) + "s, file was " + str(e.videoDuration) + "s"), "[ERROR]"
|
89 |
|
90 |
def transcribe_file(self, model: whisper.Whisper, audio_path: str, language: str, task: str = None, vad: str = None,
|
91 |
+
vadMergeWindow: float = 5, vadMaxMergeSize: float = 150, vadPadding: float = 1, vadPromptWindow: float = 1, **decodeOptions: dict):
|
92 |
# Callable for processing an audio file
|
93 |
+
whisperCallable = lambda audio, prompt : model.transcribe(audio, language=language, task=task, initial_prompt=prompt, **decodeOptions)
|
94 |
|
95 |
# The results
|
96 |
if (vad == 'silero-vad'):
|
|
|
100 |
|
101 |
process_gaps = VadSileroTranscription(transcribe_non_speech = True,
|
102 |
max_silent_period=vadMergeWindow, max_merge_size=vadMaxMergeSize,
|
103 |
+
segment_padding_left=vadPadding, segment_padding_right=vadPadding,
|
104 |
+
max_prompt_window=vadPromptWindow, copy=self.vad_model)
|
105 |
result = process_gaps.transcribe(audio_path, whisperCallable)
|
106 |
elif (vad == 'silero-vad-skip-gaps'):
|
107 |
# Use Silero VAD
|
|
|
110 |
|
111 |
skip_gaps = VadSileroTranscription(transcribe_non_speech = False,
|
112 |
max_silent_period=vadMergeWindow, max_merge_size=vadMaxMergeSize,
|
113 |
+
segment_padding_left=vadPadding, segment_padding_right=vadPadding,
|
114 |
+
max_prompt_window=vadPromptWindow, copy=self.vad_model)
|
115 |
result = skip_gaps.transcribe(audio_path, whisperCallable)
|
116 |
elif (vad == 'periodic-vad'):
|
117 |
# Very simple VAD - mark every 5 minutes as speech. This makes it less likely that Whisper enters an infinite loop, but
|
|
|
120 |
result = periodic_vad.transcribe(audio_path, whisperCallable)
|
121 |
else:
|
122 |
# Default VAD
|
123 |
+
result = whisperCallable(audio_path, None)
|
124 |
|
125 |
return result
|
126 |
|
|
|
219 |
gr.Audio(source="microphone", type="filepath", label="Microphone Input"),
|
220 |
gr.Dropdown(choices=["transcribe", "translate"], label="Task"),
|
221 |
gr.Dropdown(choices=["none", "silero-vad", "silero-vad-skip-gaps", "periodic-vad"], label="VAD"),
|
222 |
+
gr.Number(label="VAD - Merge Window (s)", precision=0, value=4),
|
223 |
+
gr.Number(label="VAD - Max Merge Size (s)", precision=0, value=30),
|
224 |
+
gr.Number(label="VAD - Padding (s)", precision=None, value=1),
|
225 |
+
gr.Number(label="VAD - Prompt Window (s)", precision=None, value=10)
|
226 |
], outputs=[
|
227 |
gr.File(label="Download"),
|
228 |
gr.Text(label="Transcription"),
|
cli.py
CHANGED
@@ -30,6 +30,7 @@ def cli():
|
|
30 |
parser.add_argument("--vad_merge_window", type=optional_float, default=5, help="The window size (in seconds) to merge voice segments")
|
31 |
parser.add_argument("--vad_max_merge_size", type=optional_float, default=150, help="The maximum size (in seconds) of a voice segment")
|
32 |
parser.add_argument("--vad_padding", type=optional_float, default=1, help="The padding (in seconds) to add to each voice segment")
|
|
|
33 |
|
34 |
parser.add_argument("--temperature", type=float, default=0, help="temperature to use for sampling")
|
35 |
parser.add_argument("--best_of", type=optional_int, default=5, help="number of candidates when sampling with non-zero temperature")
|
@@ -69,6 +70,7 @@ def cli():
|
|
69 |
vad_merge_window = args.pop("vad_merge_window")
|
70 |
vad_max_merge_size = args.pop("vad_max_merge_size")
|
71 |
vad_padding = args.pop("vad_padding")
|
|
|
72 |
|
73 |
model = whisper.load_model(model_name, device=device, download_root=model_dir)
|
74 |
transcriber = WhisperTranscriber(deleteUploadedFiles=False)
|
@@ -91,7 +93,7 @@ def cli():
|
|
91 |
|
92 |
result = transcriber.transcribe_file(model, source_path, temperature=temperature,
|
93 |
vad=vad, vadMergeWindow=vad_merge_window, vadMaxMergeSize=vad_max_merge_size,
|
94 |
-
vadPadding=vad_padding, **args)
|
95 |
|
96 |
transcriber.write_result(result, source_name, output_dir)
|
97 |
|
|
|
30 |
parser.add_argument("--vad_merge_window", type=optional_float, default=5, help="The window size (in seconds) to merge voice segments")
|
31 |
parser.add_argument("--vad_max_merge_size", type=optional_float, default=150, help="The maximum size (in seconds) of a voice segment")
|
32 |
parser.add_argument("--vad_padding", type=optional_float, default=1, help="The padding (in seconds) to add to each voice segment")
|
33 |
+
parser.add_argument("--vad_prompt_window", type=optional_float, default=0, help="The window size of the prompt to pass to Whisper")
|
34 |
|
35 |
parser.add_argument("--temperature", type=float, default=0, help="temperature to use for sampling")
|
36 |
parser.add_argument("--best_of", type=optional_int, default=5, help="number of candidates when sampling with non-zero temperature")
|
|
|
70 |
vad_merge_window = args.pop("vad_merge_window")
|
71 |
vad_max_merge_size = args.pop("vad_max_merge_size")
|
72 |
vad_padding = args.pop("vad_padding")
|
73 |
+
vad_prompt_window = args.pop("vad_prompt_window")
|
74 |
|
75 |
model = whisper.load_model(model_name, device=device, download_root=model_dir)
|
76 |
transcriber = WhisperTranscriber(deleteUploadedFiles=False)
|
|
|
93 |
|
94 |
result = transcriber.transcribe_file(model, source_path, temperature=temperature,
|
95 |
vad=vad, vadMergeWindow=vad_merge_window, vadMaxMergeSize=vad_max_merge_size,
|
96 |
+
vadPadding=vad_padding, vadPromptWindow=vad_prompt_window, **args)
|
97 |
|
98 |
transcriber.write_result(result, source_name, output_dir)
|
99 |
|
src/utils.py
CHANGED
@@ -56,7 +56,7 @@ def write_txt(transcript: Iterator[dict], file: TextIO):
|
|
56 |
def write_vtt(transcript: Iterator[dict], file: TextIO, maxLineWidth=None):
|
57 |
print("WEBVTT\n", file=file)
|
58 |
for segment in transcript:
|
59 |
-
text =
|
60 |
|
61 |
print(
|
62 |
f"{format_timestamp(segment['start'])} --> {format_timestamp(segment['end'])}\n"
|
@@ -79,7 +79,7 @@ def write_srt(transcript: Iterator[dict], file: TextIO, maxLineWidth=None):
|
|
79 |
write_srt(result["segments"], file=srt)
|
80 |
"""
|
81 |
for i, segment in enumerate(transcript, start=1):
|
82 |
-
text =
|
83 |
|
84 |
# write srt lines
|
85 |
print(
|
@@ -91,7 +91,7 @@ def write_srt(transcript: Iterator[dict], file: TextIO, maxLineWidth=None):
|
|
91 |
flush=True,
|
92 |
)
|
93 |
|
94 |
-
def
|
95 |
if (maxLineWidth is None or maxLineWidth < 0):
|
96 |
return text
|
97 |
|
|
|
56 |
def write_vtt(transcript: Iterator[dict], file: TextIO, maxLineWidth=None):
|
57 |
print("WEBVTT\n", file=file)
|
58 |
for segment in transcript:
|
59 |
+
text = process_text(segment['text'], maxLineWidth).replace('-->', '->')
|
60 |
|
61 |
print(
|
62 |
f"{format_timestamp(segment['start'])} --> {format_timestamp(segment['end'])}\n"
|
|
|
79 |
write_srt(result["segments"], file=srt)
|
80 |
"""
|
81 |
for i, segment in enumerate(transcript, start=1):
|
82 |
+
text = process_text(segment['text'].strip(), maxLineWidth).replace('-->', '->')
|
83 |
|
84 |
# write srt lines
|
85 |
print(
|
|
|
91 |
flush=True,
|
92 |
)
|
93 |
|
94 |
+
def process_text(text: str, maxLineWidth=None):
|
95 |
if (maxLineWidth is None or maxLineWidth < 0):
|
96 |
return text
|
97 |
|
src/vad.py
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
from abc import ABC, abstractmethod
|
2 |
-
from collections import Counter
|
3 |
from typing import Any, Iterator, List, Dict
|
4 |
|
5 |
from pprint import pprint
|
@@ -17,10 +17,9 @@ import ffmpeg
|
|
17 |
import numpy as np
|
18 |
|
19 |
from src.utils import format_timestamp
|
|
|
20 |
|
21 |
# Defaults for Silero
|
22 |
-
# TODO: Make these configurable?
|
23 |
-
|
24 |
SPEECH_TRESHOLD = 0.3
|
25 |
MAX_SILENT_PERIOD = 10 # seconds
|
26 |
MAX_MERGE_SIZE = 150 # Do not create segments larger than 2.5 minutes
|
@@ -35,16 +34,29 @@ TRANSCRIBE_NON_SPEECH = False
|
|
35 |
# Minimum size of segments to process
|
36 |
MIN_SEGMENT_DURATION = 1
|
37 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
VAD_MAX_PROCESSING_CHUNK = 60 * 60 # 60 minutes of audio
|
39 |
|
40 |
class AbstractTranscription(ABC):
|
41 |
-
def __init__(self, segment_padding_left: float = None, segment_padding_right = None, max_silent_period: float = None,
|
|
|
42 |
self.sampling_rate = 16000
|
43 |
self.segment_padding_left = segment_padding_left
|
44 |
self.segment_padding_right = segment_padding_right
|
45 |
self.max_silent_period = max_silent_period
|
46 |
self.max_merge_size = max_merge_size
|
47 |
self.transcribe_non_speech = transcribe_non_speech
|
|
|
|
|
|
|
|
|
48 |
|
49 |
def get_audio_segment(self, str, start_time: str = None, duration: str = None):
|
50 |
return load_audio(str, self.sampling_rate, start_time, duration)
|
@@ -74,8 +86,9 @@ class AbstractTranscription(ABC):
|
|
74 |
audio: str
|
75 |
The audio file.
|
76 |
|
77 |
-
whisperCallable: Callable[[Union[str, np.ndarray, torch.Tensor]], dict[str, Union[dict, Any]]]
|
78 |
-
The callback that is used to invoke Whisper on an audio file/buffer.
|
|
|
79 |
|
80 |
Returns
|
81 |
-------
|
@@ -86,7 +99,10 @@ class AbstractTranscription(ABC):
|
|
86 |
seconds_timestamps = self.get_transcribe_timestamps(audio)
|
87 |
|
88 |
padded = self.pad_timestamps(seconds_timestamps, self.segment_padding_left, self.segment_padding_right)
|
89 |
-
merged = self.merge_timestamps(padded, self.max_silent_period, self.max_merge_size)
|
|
|
|
|
|
|
90 |
|
91 |
print("Timestamps:")
|
92 |
pprint(merged)
|
@@ -95,7 +111,12 @@ class AbstractTranscription(ABC):
|
|
95 |
max_audio_duration = get_audio_duration(audio)
|
96 |
|
97 |
# Expand segments to include the gaps between them
|
98 |
-
|
|
|
|
|
|
|
|
|
|
|
99 |
|
100 |
print("Transcribing non-speech:")
|
101 |
pprint(merged)
|
@@ -118,10 +139,14 @@ class AbstractTranscription(ABC):
|
|
118 |
if segment_duration < MIN_SEGMENT_DURATION:
|
119 |
continue;
|
120 |
|
|
|
121 |
segment_audio = self.get_audio_segment(audio, start_time = str(segment_start), duration = str(segment_duration))
|
122 |
-
|
123 |
-
|
124 |
-
|
|
|
|
|
|
|
125 |
|
126 |
adjusted_segments = self.adjust_timestamp(segment_result["segments"], adjust_seconds=segment_start, max_source_time=segment_duration)
|
127 |
|
@@ -132,6 +157,16 @@ class AbstractTranscription(ABC):
|
|
132 |
# Increment detected language
|
133 |
languageCounter[segment_result['language']] += 1
|
134 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
135 |
if len(languageCounter) > 0:
|
136 |
result['language'] = languageCounter.most_common(1)[0][0]
|
137 |
|
@@ -203,6 +238,58 @@ class AbstractTranscription(ABC):
|
|
203 |
|
204 |
return result
|
205 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
206 |
def adjust_timestamp(self, segments: Iterator[dict], adjust_seconds: float, max_source_time: float = None):
|
207 |
result = []
|
208 |
|
@@ -253,7 +340,8 @@ class AbstractTranscription(ABC):
|
|
253 |
|
254 |
return result
|
255 |
|
256 |
-
def merge_timestamps(self, timestamps: List[Dict[str, Any]], max_merge_gap: float, max_merge_size: float
|
|
|
257 |
if max_merge_gap is None:
|
258 |
return timestamps
|
259 |
|
@@ -270,7 +358,10 @@ class AbstractTranscription(ABC):
|
|
270 |
current_entry_size = current_entry['end'] - current_entry['start']
|
271 |
|
272 |
if distance <= max_merge_gap and (max_merge_size is None or current_entry_size <= max_merge_size):
|
273 |
-
#
|
|
|
|
|
|
|
274 |
current_entry['end'] = entry['end']
|
275 |
else:
|
276 |
# Output current entry
|
@@ -299,9 +390,9 @@ class AbstractTranscription(ABC):
|
|
299 |
class VadSileroTranscription(AbstractTranscription):
|
300 |
def __init__(self, segment_padding_left=SEGMENT_PADDING_LEFT, segment_padding_right=SEGMENT_PADDING_RIGHT,
|
301 |
max_silent_period=MAX_SILENT_PERIOD, max_merge_size=MAX_MERGE_SIZE, transcribe_non_speech: bool = False,
|
302 |
-
copy = None):
|
303 |
super().__init__(segment_padding_left=segment_padding_left, segment_padding_right=segment_padding_right,
|
304 |
-
max_silent_period=max_silent_period, max_merge_size=max_merge_size, transcribe_non_speech=transcribe_non_speech)
|
305 |
|
306 |
if copy:
|
307 |
self.model = copy.model
|
|
|
1 |
from abc import ABC, abstractmethod
|
2 |
+
from collections import Counter, deque
|
3 |
from typing import Any, Iterator, List, Dict
|
4 |
|
5 |
from pprint import pprint
|
|
|
17 |
import numpy as np
|
18 |
|
19 |
from src.utils import format_timestamp
|
20 |
+
from enum import Enum
|
21 |
|
22 |
# Defaults for Silero
|
|
|
|
|
23 |
SPEECH_TRESHOLD = 0.3
|
24 |
MAX_SILENT_PERIOD = 10 # seconds
|
25 |
MAX_MERGE_SIZE = 150 # Do not create segments larger than 2.5 minutes
|
|
|
34 |
# Minimum size of segments to process
|
35 |
MIN_SEGMENT_DURATION = 1
|
36 |
|
37 |
+
# Always merge segments that are less than this duration apart
|
38 |
+
MIN_FORCE_MERGE_GAP = 0.5
|
39 |
+
FORCE_MERGE_SEGMENT_MULTIPLIER = 1.5
|
40 |
+
|
41 |
+
# The maximum time for texts from old segments to be used in the next segment
|
42 |
+
MAX_PROMPT_WINDOW = 0 # seconds (0 = disabled)
|
43 |
+
PROMPT_NO_SPEECH_PROB = 0.1 # Do not pass the text from segments with a no speech probability higher than this
|
44 |
+
|
45 |
VAD_MAX_PROCESSING_CHUNK = 60 * 60 # 60 minutes of audio
|
46 |
|
47 |
class AbstractTranscription(ABC):
|
48 |
+
def __init__(self, segment_padding_left: float = None, segment_padding_right = None, max_silent_period: float = None,
|
49 |
+
max_merge_size: float = None, transcribe_non_speech: bool = False, max_prompt_window: float = None):
|
50 |
self.sampling_rate = 16000
|
51 |
self.segment_padding_left = segment_padding_left
|
52 |
self.segment_padding_right = segment_padding_right
|
53 |
self.max_silent_period = max_silent_period
|
54 |
self.max_merge_size = max_merge_size
|
55 |
self.transcribe_non_speech = transcribe_non_speech
|
56 |
+
self.max_prompt_window = max_prompt_window
|
57 |
+
|
58 |
+
self.min_force_merge_gap = MIN_FORCE_MERGE_GAP
|
59 |
+
self.max_force_merge_size = max_merge_size * FORCE_MERGE_SEGMENT_MULTIPLIER if max_merge_size is not None else None
|
60 |
|
61 |
def get_audio_segment(self, str, start_time: str = None, duration: str = None):
|
62 |
return load_audio(str, self.sampling_rate, start_time, duration)
|
|
|
86 |
audio: str
|
87 |
The audio file.
|
88 |
|
89 |
+
whisperCallable: Callable[[Union[str, np.ndarray, torch.Tensor], str], dict[str, Union[dict, Any]]]
|
90 |
+
The callback that is used to invoke Whisper on an audio file/buffer. The first parameter is the audio file/buffer,
|
91 |
+
and the second parameter is an optional text prompt. The return value is the result of the Whisper call.
|
92 |
|
93 |
Returns
|
94 |
-------
|
|
|
99 |
seconds_timestamps = self.get_transcribe_timestamps(audio)
|
100 |
|
101 |
padded = self.pad_timestamps(seconds_timestamps, self.segment_padding_left, self.segment_padding_right)
|
102 |
+
merged = self.merge_timestamps(padded, self.max_silent_period, self.max_merge_size, self.min_force_merge_gap, self.max_force_merge_size)
|
103 |
+
|
104 |
+
# A deque of transcribed segments that is passed to the next segment as a prompt
|
105 |
+
prompt_window = deque()
|
106 |
|
107 |
print("Timestamps:")
|
108 |
pprint(merged)
|
|
|
111 |
max_audio_duration = get_audio_duration(audio)
|
112 |
|
113 |
# Expand segments to include the gaps between them
|
114 |
+
if (self.max_prompt_window is not None and self.max_prompt_window > 0):
|
115 |
+
# When we have a prompt window, we create speech segments betwen each segment if we exceed the merge size
|
116 |
+
merged = self.fill_gaps(merged, total_duration=max_audio_duration, max_expand_size=self.max_merge_size)
|
117 |
+
else:
|
118 |
+
# With no prompt window, it is better to expand the segments
|
119 |
+
merged = self.expand_gaps(merged, total_duration=max_audio_duration)
|
120 |
|
121 |
print("Transcribing non-speech:")
|
122 |
pprint(merged)
|
|
|
139 |
if segment_duration < MIN_SEGMENT_DURATION:
|
140 |
continue;
|
141 |
|
142 |
+
# Audio to run on Whisper
|
143 |
segment_audio = self.get_audio_segment(audio, start_time = str(segment_start), duration = str(segment_duration))
|
144 |
+
# Previous segments to use as a prompt
|
145 |
+
segment_prompt = ' '.join([segment['text'] for segment in prompt_window]) if len(prompt_window) > 0 else None
|
146 |
+
|
147 |
+
print("Running whisper from ", format_timestamp(segment_start), " to ", format_timestamp(segment_end), ", duration: ",
|
148 |
+
segment_duration, "expanded: ", segment_expand_amount, "prompt: ", segment_prompt)
|
149 |
+
segment_result = whisperCallable(segment_audio, segment_prompt)
|
150 |
|
151 |
adjusted_segments = self.adjust_timestamp(segment_result["segments"], adjust_seconds=segment_start, max_source_time=segment_duration)
|
152 |
|
|
|
157 |
# Increment detected language
|
158 |
languageCounter[segment_result['language']] += 1
|
159 |
|
160 |
+
# Update prompt window
|
161 |
+
if (self.max_prompt_window is not None and self.max_prompt_window > 0):
|
162 |
+
# Add segments to the current prompt window
|
163 |
+
for segment in adjusted_segments:
|
164 |
+
if segment.get('no_speech_prob', 0) <= PROMPT_NO_SPEECH_PROB:
|
165 |
+
prompt_window.append(segment)
|
166 |
+
|
167 |
+
while (len(prompt_window) > 0 and prompt_window[0]['end'] < segment_end - self.max_prompt_window):
|
168 |
+
prompt_window.popleft()
|
169 |
+
|
170 |
if len(languageCounter) > 0:
|
171 |
result['language'] = languageCounter.most_common(1)[0][0]
|
172 |
|
|
|
238 |
|
239 |
return result
|
240 |
|
241 |
+
def fill_gaps(self, segments: List[Dict[str, Any]], total_duration: float, max_expand_size: float = None):
|
242 |
+
result = []
|
243 |
+
|
244 |
+
if len(segments) == 0:
|
245 |
+
return result
|
246 |
+
|
247 |
+
# Add gap at the beginning if needed
|
248 |
+
if (segments[0]['start'] > 0):
|
249 |
+
result.append({ 'start': 0, 'end': segments[0]['start'], 'gap': True } )
|
250 |
+
|
251 |
+
for i in range(len(segments) - 1):
|
252 |
+
expanded = False
|
253 |
+
current_segment = segments[i]
|
254 |
+
next_segment = segments[i + 1]
|
255 |
+
|
256 |
+
delta = next_segment['start'] - current_segment['end']
|
257 |
+
|
258 |
+
if (max_expand_size is not None and delta <= max_expand_size):
|
259 |
+
# Just expand the current segment
|
260 |
+
current_segment = current_segment.copy()
|
261 |
+
current_segment['expand_amount'] = delta
|
262 |
+
current_segment['end'] = next_segment['start']
|
263 |
+
expanded = True
|
264 |
+
|
265 |
+
result.append(current_segment)
|
266 |
+
|
267 |
+
# Add a gap to the next segment if needed
|
268 |
+
if (delta >= 0 and not expanded):
|
269 |
+
result.append({ 'start': current_segment['end'], 'end': next_segment['start'], 'gap': True } )
|
270 |
+
|
271 |
+
# Add last segment
|
272 |
+
last_segment = segments[-1]
|
273 |
+
result.append(last_segment)
|
274 |
+
|
275 |
+
# Also include total duration if specified
|
276 |
+
if (total_duration is not None):
|
277 |
+
last_segment = result[-1]
|
278 |
+
|
279 |
+
delta = total_duration - last_segment['end']
|
280 |
+
|
281 |
+
if (delta > 0):
|
282 |
+
if (max_expand_size is not None and delta <= max_expand_size):
|
283 |
+
# Expand the last segment
|
284 |
+
last_segment = last_segment.copy()
|
285 |
+
last_segment['expand_amount'] = delta
|
286 |
+
last_segment['end'] = total_duration
|
287 |
+
result[-1] = last_segment
|
288 |
+
else:
|
289 |
+
result.append({ 'start': last_segment['end'], 'end': total_duration, 'gap': True } )
|
290 |
+
|
291 |
+
return result
|
292 |
+
|
293 |
def adjust_timestamp(self, segments: Iterator[dict], adjust_seconds: float, max_source_time: float = None):
|
294 |
result = []
|
295 |
|
|
|
340 |
|
341 |
return result
|
342 |
|
343 |
+
def merge_timestamps(self, timestamps: List[Dict[str, Any]], max_merge_gap: float, max_merge_size: float,
|
344 |
+
min_force_merge_gap: float, max_force_merge_size: float):
|
345 |
if max_merge_gap is None:
|
346 |
return timestamps
|
347 |
|
|
|
358 |
current_entry_size = current_entry['end'] - current_entry['start']
|
359 |
|
360 |
if distance <= max_merge_gap and (max_merge_size is None or current_entry_size <= max_merge_size):
|
361 |
+
# Regular merge
|
362 |
+
current_entry['end'] = entry['end']
|
363 |
+
elif min_force_merge_gap is not None and distance <= min_force_merge_gap and (max_force_merge_size is None or current_entry_size <= max_force_merge_size):
|
364 |
+
# Force merge if the distance is small (up to a certain maximum size)
|
365 |
current_entry['end'] = entry['end']
|
366 |
else:
|
367 |
# Output current entry
|
|
|
390 |
class VadSileroTranscription(AbstractTranscription):
|
391 |
def __init__(self, segment_padding_left=SEGMENT_PADDING_LEFT, segment_padding_right=SEGMENT_PADDING_RIGHT,
|
392 |
max_silent_period=MAX_SILENT_PERIOD, max_merge_size=MAX_MERGE_SIZE, transcribe_non_speech: bool = False,
|
393 |
+
max_prompt_window=MAX_PROMPT_WINDOW, copy = None):
|
394 |
super().__init__(segment_padding_left=segment_padding_left, segment_padding_right=segment_padding_right,
|
395 |
+
max_silent_period=max_silent_period, max_merge_size=max_merge_size, transcribe_non_speech=transcribe_non_speech, max_prompt_window=max_prompt_window)
|
396 |
|
397 |
if copy:
|
398 |
self.model = copy.model
|