Spaces:
Sleeping
Sleeping
Update helper_fns.py
Browse files- helper_fns.py +52 -29
helper_fns.py
CHANGED
|
@@ -1,30 +1,53 @@
|
|
| 1 |
-
import gradio as gr
|
| 2 |
-
from transformers import pipeline
|
| 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 |
return audio_path
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
from transformers import pipeline
|
| 3 |
+
from gtts import gTTS
|
| 4 |
+
from pydub import AudioSegment
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
#text to sppech function
|
| 8 |
+
def text_to_speech(text):
|
| 9 |
+
# Convert text to speech with a US accent using gTTS
|
| 10 |
+
tts = gTTS(text=text, lang='en', tld='us', slow=False)
|
| 11 |
+
tts.save('temp.mp3')
|
| 12 |
+
|
| 13 |
+
# Load the audio file
|
| 14 |
+
audio = AudioSegment.from_file('temp.mp3')
|
| 15 |
+
|
| 16 |
+
# Adjust the speed to approximately 170 wpm
|
| 17 |
+
playback_speed = 1.20
|
| 18 |
+
audio = audio.speedup(playback_speed=playback_speed)
|
| 19 |
+
|
| 20 |
+
# Save and return the adjusted audio file
|
| 21 |
+
final_filename = 'text_to_speech.mp3'
|
| 22 |
+
audio.export(final_filename, format='mp3')
|
| 23 |
+
|
| 24 |
+
return final_filename
|
| 25 |
+
|
| 26 |
+
|
| 27 |
+
def process_files():
|
| 28 |
+
return (gr.update(interactive=True,
|
| 29 |
+
elem_id='summary_button'),
|
| 30 |
+
gr.update(interactive = True, elem_id = 'summarization_method')
|
| 31 |
+
)
|
| 32 |
+
|
| 33 |
+
|
| 34 |
+
|
| 35 |
+
def get_summarization_method(option):
|
| 36 |
+
return option
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
|
| 41 |
+
def text_to_audio(text, model_name="facebook/fastspeech2-en-ljspeech"):
|
| 42 |
+
# Initialize the TTS pipeline
|
| 43 |
+
tts_pipeline = pipeline("text-to-speech", model=model_name)
|
| 44 |
+
|
| 45 |
+
# Generate the audio from text
|
| 46 |
+
audio = tts_pipeline(text)
|
| 47 |
+
|
| 48 |
+
# Save the audio to a file
|
| 49 |
+
audio_path = "output.wav"
|
| 50 |
+
with open(audio_path, "wb") as file:
|
| 51 |
+
file.write(audio["wav"])
|
| 52 |
+
|
| 53 |
return audio_path
|