Spaces:
Runtime error
Runtime error
Finalize HF demo
Browse filesSigned-off-by: smajumdar <[email protected]>
app.py
CHANGED
@@ -4,23 +4,37 @@ import uuid
|
|
4 |
import tempfile
|
5 |
import subprocess
|
6 |
import re
|
|
|
7 |
|
8 |
import gradio as gr
|
9 |
import pytube as pt
|
10 |
|
11 |
import nemo.collections.asr as nemo_asr
|
|
|
|
|
12 |
import speech_to_text_buffered_infer_ctc as buffered_ctc
|
13 |
import speech_to_text_buffered_infer_rnnt as buffered_rnnt
|
|
|
14 |
|
15 |
# Set NeMo cache dir as /tmp
|
16 |
from nemo import constants
|
17 |
-
os.environ[constants.NEMO_ENV_CACHE_DIR] = "/tmp/nemo"
|
18 |
|
|
|
|
|
|
|
|
|
|
|
19 |
|
20 |
-
SAMPLE_RATE = 16000
|
21 |
TITLE = "NeMo ASR Inference on Hugging Face"
|
22 |
DESCRIPTION = "Demo of all languages supported by NeMo ASR"
|
23 |
DEFAULT_EN_MODEL = "nvidia/stt_en_conformer_transducer_xlarge"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
|
25 |
MARKDOWN = f"""
|
26 |
# {TITLE}
|
@@ -32,6 +46,13 @@ CSS = """
|
|
32 |
p.big {
|
33 |
font-size: 20px;
|
34 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
"""
|
36 |
|
37 |
ARTICLE = """
|
@@ -58,6 +79,9 @@ for info in hf_infos:
|
|
58 |
|
59 |
SUPPORTED_MODEL_NAMES = sorted(list(SUPPORTED_MODEL_NAMES))
|
60 |
|
|
|
|
|
|
|
61 |
model_dict = {model_name: gr.Interface.load(f'models/{model_name}') for model_name in SUPPORTED_MODEL_NAMES}
|
62 |
|
63 |
SUPPORTED_LANG_MODEL_DICT = {}
|
@@ -77,6 +101,14 @@ for lang in SUPPORTED_LANG_MODEL_DICT.keys():
|
|
77 |
SUPPORTED_LANG_MODEL_DICT[lang] = model_ids
|
78 |
|
79 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
80 |
def parse_duration(audio_file):
|
81 |
"""
|
82 |
FFMPEG to calculate durations. Libraries can do it too, but filetypes cause different libraries to behave differently.
|
@@ -108,7 +140,7 @@ def resolve_model_type(model_name: str) -> str:
|
|
108 |
return 'ctc'
|
109 |
|
110 |
# Model specific maps
|
111 |
-
|
112 |
return 'ctc'
|
113 |
elif 'quartznet' in model_name:
|
114 |
return 'ctc'
|
@@ -116,9 +148,8 @@ def resolve_model_type(model_name: str) -> str:
|
|
116 |
return 'ctc'
|
117 |
elif 'contextnet' in model_name:
|
118 |
return 'ctc'
|
119 |
-
|
120 |
-
|
121 |
-
return None
|
122 |
|
123 |
|
124 |
def resolve_model_stride(model_name) -> int:
|
@@ -185,6 +216,16 @@ def extract_result_from_manifest(filepath, model_name) -> (bool, str):
|
|
185 |
return False, f"Could not perform inference on model with name : {model_name}"
|
186 |
|
187 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
188 |
def infer_audio(model_name: str, audio_file: str) -> str:
|
189 |
"""
|
190 |
Main method that switches from HF inference for small audio files to Buffered CTC/RNNT mode for long audio files.
|
@@ -195,17 +236,18 @@ def infer_audio(model_name: str, audio_file: str) -> str:
|
|
195 |
|
196 |
Returns:
|
197 |
str which is the transcription if successful.
|
|
|
198 |
"""
|
199 |
# Parse the duration of the audio file
|
200 |
duration = parse_duration(audio_file)
|
201 |
|
202 |
-
if duration >
|
203 |
# Process audio to be of wav type (possible youtube audio)
|
204 |
audio_file = convert_audio(audio_file)
|
205 |
|
206 |
# If audio file transcoding failed, let user know
|
207 |
if audio_file is None:
|
208 |
-
return "Failed to convert audio file to wav."
|
209 |
|
210 |
# Extract audio dir from resolved audio filepath
|
211 |
audio_dir = os.path.split(audio_file)[0]
|
@@ -214,7 +256,7 @@ def infer_audio(model_name: str, audio_file: str) -> str:
|
|
214 |
model_stride = resolve_model_stride(model_name)
|
215 |
|
216 |
if model_stride < 0:
|
217 |
-
return f"Failed to compute the model stride for model with name : {model_name}"
|
218 |
|
219 |
# Process model type (CTC/RNNT/Hybrid)
|
220 |
model_type = resolve_model_type(model_name)
|
@@ -266,7 +308,7 @@ def infer_audio(model_name: str, audio_file: str) -> str:
|
|
266 |
pass
|
267 |
|
268 |
if RESULT is None:
|
269 |
-
return f"Could not parse model type; failed to perform inference with model {model_name}!"
|
270 |
|
271 |
elif model_type == 'ctc':
|
272 |
|
@@ -303,9 +345,10 @@ def infer_audio(model_name: str, audio_file: str) -> str:
|
|
303 |
return extract_result_from_manifest('output.json', model_name)[-1]
|
304 |
|
305 |
else:
|
306 |
-
return f"Could not parse model type; failed to perform inference with model {model_name}!"
|
307 |
|
308 |
else:
|
|
|
309 |
if model_name in model_dict:
|
310 |
model = model_dict[model_name]
|
311 |
else:
|
@@ -317,7 +360,7 @@ def infer_audio(model_name: str, audio_file: str) -> str:
|
|
317 |
return transcriptions
|
318 |
else:
|
319 |
error = (
|
320 |
-
f"Could not find model {model_name} in list of available models : "
|
321 |
f"{list([k for k in model_dict.keys()])}"
|
322 |
)
|
323 |
return error
|
@@ -334,30 +377,60 @@ def transcribe(microphone, audio_file, model_name):
|
|
334 |
audio_data = microphone
|
335 |
|
336 |
elif (microphone is None) and (audio_file is None):
|
337 |
-
|
338 |
|
339 |
elif microphone is not None:
|
340 |
audio_data = microphone
|
341 |
else:
|
342 |
audio_data = audio_file
|
343 |
|
|
|
344 |
try:
|
345 |
# Use HF API for transcription
|
|
|
346 |
transcriptions = infer_audio(model_name, audio_data)
|
|
|
|
|
347 |
|
348 |
except Exception as e:
|
349 |
transcriptions = ""
|
350 |
-
warn_output = warn_output
|
|
|
|
|
|
|
|
|
351 |
warn_output += (
|
352 |
f"The model `{model_name}` is currently loading and cannot be used "
|
353 |
-
f"for transcription
|
354 |
f"Please try another model or wait a few minutes."
|
355 |
)
|
356 |
|
357 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
358 |
|
359 |
|
360 |
def _return_yt_html_embed(yt_url):
|
|
|
361 |
video_id = yt_url.split("?v=")[-1]
|
362 |
HTML_str = (
|
363 |
f'<center> <iframe width="500" height="320" src="https://www.youtube.com/embed/{video_id}"> </iframe>'
|
@@ -367,6 +440,7 @@ def _return_yt_html_embed(yt_url):
|
|
367 |
|
368 |
|
369 |
def yt_transcribe(yt_url, model_name):
|
|
|
370 |
yt = pt.YouTube(yt_url)
|
371 |
html_embed_str = _return_yt_html_embed(yt_url)
|
372 |
|
@@ -374,15 +448,57 @@ def yt_transcribe(yt_url, model_name):
|
|
374 |
file_uuid = str(uuid.uuid4().hex)
|
375 |
file_uuid = f"{tempdir}/{file_uuid}.mp3"
|
376 |
|
|
|
|
|
|
|
377 |
stream = yt.streams.filter(only_audio=True)[0]
|
378 |
stream.download(filename=file_uuid)
|
379 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
380 |
text = infer_audio(model_name, file_uuid)
|
381 |
|
382 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
383 |
|
384 |
|
385 |
def create_lang_selector_component(default_en_model=DEFAULT_EN_MODEL):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
386 |
lang_selector = gr.components.Dropdown(
|
387 |
choices=sorted(list(SUPPORTED_LANGUAGES)), value="en", type="value", label="Languages", interactive=True,
|
388 |
)
|
@@ -406,6 +522,9 @@ def create_lang_selector_component(default_en_model=DEFAULT_EN_MODEL):
|
|
406 |
return lang_selector, models_in_lang
|
407 |
|
408 |
|
|
|
|
|
|
|
409 |
demo = gr.Blocks(title=TITLE, css=CSS)
|
410 |
|
411 |
with demo:
|
@@ -419,9 +538,12 @@ with demo:
|
|
419 |
lang_selector, models_in_lang = create_lang_selector_component()
|
420 |
|
421 |
transcript = gr.components.Label(label='Transcript')
|
|
|
422 |
|
423 |
run = gr.components.Button('Transcribe')
|
424 |
-
run.click(
|
|
|
|
|
425 |
|
426 |
with gr.Tab("Transcribe Youtube"):
|
427 |
yt_url = gr.components.Textbox(
|
@@ -429,14 +551,19 @@ with demo:
|
|
429 |
)
|
430 |
|
431 |
lang_selector_yt, models_in_lang_yt = create_lang_selector_component(
|
432 |
-
default_en_model=
|
433 |
)
|
434 |
|
435 |
-
|
|
|
|
|
|
|
436 |
transcript = gr.components.Label(label='Transcript')
|
|
|
437 |
|
438 |
-
run
|
439 |
-
|
|
|
440 |
|
441 |
gr.components.HTML(ARTICLE)
|
442 |
|
|
|
4 |
import tempfile
|
5 |
import subprocess
|
6 |
import re
|
7 |
+
import time
|
8 |
|
9 |
import gradio as gr
|
10 |
import pytube as pt
|
11 |
|
12 |
import nemo.collections.asr as nemo_asr
|
13 |
+
import torch
|
14 |
+
|
15 |
import speech_to_text_buffered_infer_ctc as buffered_ctc
|
16 |
import speech_to_text_buffered_infer_rnnt as buffered_rnnt
|
17 |
+
from nemo.utils import logging
|
18 |
|
19 |
# Set NeMo cache dir as /tmp
|
20 |
from nemo import constants
|
|
|
21 |
|
22 |
+
os.environ[constants.NEMO_ENV_CACHE_DIR] = "/tmp/nemo/"
|
23 |
+
|
24 |
+
|
25 |
+
SAMPLE_RATE = 16000 # Default sample rate for ASR
|
26 |
+
BUFFERED_INFERENCE_DURATION_THRESHOLD = 60.0 # 60 second and above will require chunked inference.
|
27 |
|
|
|
28 |
TITLE = "NeMo ASR Inference on Hugging Face"
|
29 |
DESCRIPTION = "Demo of all languages supported by NeMo ASR"
|
30 |
DEFAULT_EN_MODEL = "nvidia/stt_en_conformer_transducer_xlarge"
|
31 |
+
DEFAULT_BUFFERED_EN_MODEL = "nvidia/stt_en_conformer_transducer_large"
|
32 |
+
|
33 |
+
# Pre-download and cache the model in disk space
|
34 |
+
logging.setLevel(logging.ERROR)
|
35 |
+
tmp_model = nemo_asr.models.ASRModel.from_pretrained(DEFAULT_BUFFERED_EN_MODEL, map_location='cpu')
|
36 |
+
del tmp_model
|
37 |
+
logging.setLevel(logging.INFO)
|
38 |
|
39 |
MARKDOWN = f"""
|
40 |
# {TITLE}
|
|
|
46 |
p.big {
|
47 |
font-size: 20px;
|
48 |
}
|
49 |
+
|
50 |
+
/* From https://huggingface.co/spaces/k2-fsa/automatic-speech-recognition/blob/main/app.py */
|
51 |
+
|
52 |
+
.result {display:flex;flex-direction:column}
|
53 |
+
.result_item {padding:15px;margin-bottom:8px;border-radius:15px;width:100%;font-size:20px;}
|
54 |
+
.result_item_success {background-color:mediumaquamarine;color:white;align-self:start}
|
55 |
+
.result_item_error {background-color:#ff7070;color:white;align-self:start}
|
56 |
"""
|
57 |
|
58 |
ARTICLE = """
|
|
|
79 |
|
80 |
SUPPORTED_MODEL_NAMES = sorted(list(SUPPORTED_MODEL_NAMES))
|
81 |
|
82 |
+
# DEBUG FILTER
|
83 |
+
SUPPORTED_MODEL_NAMES = list(filter(lambda x: "en" in x and "conformer_transducer_large" in x, SUPPORTED_MODEL_NAMES))
|
84 |
+
|
85 |
model_dict = {model_name: gr.Interface.load(f'models/{model_name}') for model_name in SUPPORTED_MODEL_NAMES}
|
86 |
|
87 |
SUPPORTED_LANG_MODEL_DICT = {}
|
|
|
101 |
SUPPORTED_LANG_MODEL_DICT[lang] = model_ids
|
102 |
|
103 |
|
104 |
+
def get_device():
|
105 |
+
gpu_available = torch.cuda.is_available()
|
106 |
+
if gpu_available:
|
107 |
+
return torch.cuda.get_device_name()
|
108 |
+
else:
|
109 |
+
return "CPU"
|
110 |
+
|
111 |
+
|
112 |
def parse_duration(audio_file):
|
113 |
"""
|
114 |
FFMPEG to calculate durations. Libraries can do it too, but filetypes cause different libraries to behave differently.
|
|
|
140 |
return 'ctc'
|
141 |
|
142 |
# Model specific maps
|
143 |
+
if 'jasper' in model_name:
|
144 |
return 'ctc'
|
145 |
elif 'quartznet' in model_name:
|
146 |
return 'ctc'
|
|
|
148 |
return 'ctc'
|
149 |
elif 'contextnet' in model_name:
|
150 |
return 'ctc'
|
151 |
+
|
152 |
+
return None
|
|
|
153 |
|
154 |
|
155 |
def resolve_model_stride(model_name) -> int:
|
|
|
216 |
return False, f"Could not perform inference on model with name : {model_name}"
|
217 |
|
218 |
|
219 |
+
def build_html_output(s: str, style: str = "result_item_success"):
|
220 |
+
return f"""
|
221 |
+
<div class='result'>
|
222 |
+
<div class='result_item {style}'>
|
223 |
+
{s}
|
224 |
+
</div>
|
225 |
+
</div>
|
226 |
+
"""
|
227 |
+
|
228 |
+
|
229 |
def infer_audio(model_name: str, audio_file: str) -> str:
|
230 |
"""
|
231 |
Main method that switches from HF inference for small audio files to Buffered CTC/RNNT mode for long audio files.
|
|
|
236 |
|
237 |
Returns:
|
238 |
str which is the transcription if successful.
|
239 |
+
str which is HTML output of logs.
|
240 |
"""
|
241 |
# Parse the duration of the audio file
|
242 |
duration = parse_duration(audio_file)
|
243 |
|
244 |
+
if duration > BUFFERED_INFERENCE_DURATION_THRESHOLD: # Longer than one minute; use buffered mode
|
245 |
# Process audio to be of wav type (possible youtube audio)
|
246 |
audio_file = convert_audio(audio_file)
|
247 |
|
248 |
# If audio file transcoding failed, let user know
|
249 |
if audio_file is None:
|
250 |
+
return "Error:- Failed to convert audio file to wav."
|
251 |
|
252 |
# Extract audio dir from resolved audio filepath
|
253 |
audio_dir = os.path.split(audio_file)[0]
|
|
|
256 |
model_stride = resolve_model_stride(model_name)
|
257 |
|
258 |
if model_stride < 0:
|
259 |
+
return f"Error:- Failed to compute the model stride for model with name : {model_name}"
|
260 |
|
261 |
# Process model type (CTC/RNNT/Hybrid)
|
262 |
model_type = resolve_model_type(model_name)
|
|
|
308 |
pass
|
309 |
|
310 |
if RESULT is None:
|
311 |
+
return f"Error:- Could not parse model type; failed to perform inference with model {model_name}!"
|
312 |
|
313 |
elif model_type == 'ctc':
|
314 |
|
|
|
345 |
return extract_result_from_manifest('output.json', model_name)[-1]
|
346 |
|
347 |
else:
|
348 |
+
return f"Error:- Could not parse model type; failed to perform inference with model {model_name}!"
|
349 |
|
350 |
else:
|
351 |
+
# Obtain Gradio Model function from cache of models
|
352 |
if model_name in model_dict:
|
353 |
model = model_dict[model_name]
|
354 |
else:
|
|
|
360 |
return transcriptions
|
361 |
else:
|
362 |
error = (
|
363 |
+
f"Error:- Could not find model {model_name} in list of available models : "
|
364 |
f"{list([k for k in model_dict.keys()])}"
|
365 |
)
|
366 |
return error
|
|
|
377 |
audio_data = microphone
|
378 |
|
379 |
elif (microphone is None) and (audio_file is None):
|
380 |
+
warn_output = "ERROR: You have to either use the microphone or upload an audio file"
|
381 |
|
382 |
elif microphone is not None:
|
383 |
audio_data = microphone
|
384 |
else:
|
385 |
audio_data = audio_file
|
386 |
|
387 |
+
time_diff = None
|
388 |
try:
|
389 |
# Use HF API for transcription
|
390 |
+
start = time.time()
|
391 |
transcriptions = infer_audio(model_name, audio_data)
|
392 |
+
end = time.time()
|
393 |
+
time_diff = end - start
|
394 |
|
395 |
except Exception as e:
|
396 |
transcriptions = ""
|
397 |
+
warn_output = warn_output
|
398 |
+
|
399 |
+
if warn_output != "":
|
400 |
+
warn_output += "<br><br>"
|
401 |
+
|
402 |
warn_output += (
|
403 |
f"The model `{model_name}` is currently loading and cannot be used "
|
404 |
+
f"for transcription.<br>"
|
405 |
f"Please try another model or wait a few minutes."
|
406 |
)
|
407 |
|
408 |
+
# Built HTML output
|
409 |
+
if warn_output != "":
|
410 |
+
html_output = build_html_output(warn_output, style="result_item_error")
|
411 |
+
else:
|
412 |
+
if transcriptions.startswith("Error:-"):
|
413 |
+
html_output = build_html_output(transcriptions, style="result_item_error")
|
414 |
+
else:
|
415 |
+
audio_duration = parse_duration(audio_data)
|
416 |
+
|
417 |
+
output = f"Successfully transcribed on {get_device()} ! <br>" f"Transcription Time : {time_diff: 0.3f} s"
|
418 |
+
|
419 |
+
if audio_duration > BUFFERED_INFERENCE_DURATION_THRESHOLD:
|
420 |
+
output += f""" <br><br>
|
421 |
+
Note: Audio duration was {audio_duration: 0.3f} s, so model had to be downloaded, initialized, and then
|
422 |
+
buffered inference was used. <br>
|
423 |
+
|
424 |
+
Please rerun again in order to measure the time taken for just inference with pre-downloaded model. <br>
|
425 |
+
"""
|
426 |
+
|
427 |
+
html_output = build_html_output(output)
|
428 |
+
|
429 |
+
return transcriptions, html_output
|
430 |
|
431 |
|
432 |
def _return_yt_html_embed(yt_url):
|
433 |
+
""" Obtained from https://huggingface.co/spaces/whisper-event/whisper-demo """
|
434 |
video_id = yt_url.split("?v=")[-1]
|
435 |
HTML_str = (
|
436 |
f'<center> <iframe width="500" height="320" src="https://www.youtube.com/embed/{video_id}"> </iframe>'
|
|
|
440 |
|
441 |
|
442 |
def yt_transcribe(yt_url, model_name):
|
443 |
+
""" Modified from https://huggingface.co/spaces/whisper-event/whisper-demo """
|
444 |
yt = pt.YouTube(yt_url)
|
445 |
html_embed_str = _return_yt_html_embed(yt_url)
|
446 |
|
|
|
448 |
file_uuid = str(uuid.uuid4().hex)
|
449 |
file_uuid = f"{tempdir}/{file_uuid}.mp3"
|
450 |
|
451 |
+
# Download YT Audio temporarily
|
452 |
+
download_time_start = time.time()
|
453 |
+
|
454 |
stream = yt.streams.filter(only_audio=True)[0]
|
455 |
stream.download(filename=file_uuid)
|
456 |
|
457 |
+
download_time_end = time.time()
|
458 |
+
|
459 |
+
# Get audio duration
|
460 |
+
audio_duration = parse_duration(file_uuid)
|
461 |
+
|
462 |
+
# Perform transcription
|
463 |
+
infer_time_start = time.time()
|
464 |
+
|
465 |
text = infer_audio(model_name, file_uuid)
|
466 |
|
467 |
+
infer_time_end = time.time()
|
468 |
+
|
469 |
+
if text.startswith("Error:-"):
|
470 |
+
html_output = build_html_output(text, style='result_item_error')
|
471 |
+
else:
|
472 |
+
html_output = f"""
|
473 |
+
Successfully transcribed on {get_device()} ! <br>
|
474 |
+
Audio Download Time : {download_time_end - download_time_start: 0.3f} s <br>
|
475 |
+
Transcription Time : {infer_time_end - infer_time_start: 0.3f} s <br>
|
476 |
+
"""
|
477 |
+
|
478 |
+
if audio_duration > BUFFERED_INFERENCE_DURATION_THRESHOLD:
|
479 |
+
html_output += f""" <br>
|
480 |
+
Note: Audio duration was {audio_duration: 0.3f} s, so model had to be downloaded, initialized, and then
|
481 |
+
buffered inference was used. <br>
|
482 |
+
|
483 |
+
Please rerun again in order to measure the time taken for just inference with pre-downloaded model. <br>
|
484 |
+
"""
|
485 |
+
|
486 |
+
html_output = build_html_output(html_output)
|
487 |
+
|
488 |
+
return text, html_embed_str, html_output
|
489 |
|
490 |
|
491 |
def create_lang_selector_component(default_en_model=DEFAULT_EN_MODEL):
|
492 |
+
"""
|
493 |
+
Utility function to select a langauge from a dropdown menu, and simultanously update another dropdown
|
494 |
+
containing the corresponding model checkpoints for that language.
|
495 |
+
|
496 |
+
Args:
|
497 |
+
default_en_model: str name of a default english model that should be the set default.
|
498 |
+
|
499 |
+
Returns:
|
500 |
+
Gradio components for lang_selector (Dropdown menu) and models_in_lang (Dropdown menu)
|
501 |
+
"""
|
502 |
lang_selector = gr.components.Dropdown(
|
503 |
choices=sorted(list(SUPPORTED_LANGUAGES)), value="en", type="value", label="Languages", interactive=True,
|
504 |
)
|
|
|
522 |
return lang_selector, models_in_lang
|
523 |
|
524 |
|
525 |
+
"""
|
526 |
+
Define the GUI
|
527 |
+
"""
|
528 |
demo = gr.Blocks(title=TITLE, css=CSS)
|
529 |
|
530 |
with demo:
|
|
|
538 |
lang_selector, models_in_lang = create_lang_selector_component()
|
539 |
|
540 |
transcript = gr.components.Label(label='Transcript')
|
541 |
+
audio_html_output = gr.components.HTML()
|
542 |
|
543 |
run = gr.components.Button('Transcribe')
|
544 |
+
run.click(
|
545 |
+
transcribe, inputs=[microphone, file_upload, models_in_lang], outputs=[transcript, audio_html_output]
|
546 |
+
)
|
547 |
|
548 |
with gr.Tab("Transcribe Youtube"):
|
549 |
yt_url = gr.components.Textbox(
|
|
|
551 |
)
|
552 |
|
553 |
lang_selector_yt, models_in_lang_yt = create_lang_selector_component(
|
554 |
+
default_en_model=DEFAULT_BUFFERED_EN_MODEL
|
555 |
)
|
556 |
|
557 |
+
with gr.Row():
|
558 |
+
run = gr.components.Button('Transcribe YouTube')
|
559 |
+
embedded_video = gr.components.HTML()
|
560 |
+
|
561 |
transcript = gr.components.Label(label='Transcript')
|
562 |
+
yt_html_output = gr.components.HTML()
|
563 |
|
564 |
+
run.click(
|
565 |
+
yt_transcribe, inputs=[yt_url, models_in_lang_yt], outputs=[transcript, embedded_video, yt_html_output]
|
566 |
+
)
|
567 |
|
568 |
gr.components.HTML(ARTICLE)
|
569 |
|