Spaces:
Sleeping
Sleeping
Ervinoreo
commited on
Commit
·
b9d31ad
1
Parent(s):
d8cd76e
Add application file
Browse files
app.py
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
import speech_recognition as sr
|
| 3 |
+
from deep_translator import GoogleTranslator
|
| 4 |
+
from gtts import gTTS
|
| 5 |
+
import os
|
| 6 |
+
|
| 7 |
+
# Recognize and translate audio to Chinese
|
| 8 |
+
def recognize_and_translate(audio_file_path):
|
| 9 |
+
# Use the file path directly from Gradio input
|
| 10 |
+
recognizer = sr.Recognizer()
|
| 11 |
+
with sr.AudioFile(audio_file_path) as source:
|
| 12 |
+
audio_data = recognizer.record(source)
|
| 13 |
+
try:
|
| 14 |
+
text = recognizer.recognize_google(audio_data)
|
| 15 |
+
except sr.UnknownValueError:
|
| 16 |
+
return "Could not understand the audio."
|
| 17 |
+
except sr.RequestError:
|
| 18 |
+
return "Network error. Please try again."
|
| 19 |
+
|
| 20 |
+
# Translate the recognized text to Chinese
|
| 21 |
+
translated_text = GoogleTranslator(source="en", target="zh-CN").translate(text)
|
| 22 |
+
return f"Recognized: {text}\nTranslated to Chinese: {translated_text}"
|
| 23 |
+
|
| 24 |
+
# Translate Chinese input to English and convert to speech
|
| 25 |
+
def translate_to_english(chinese_text):
|
| 26 |
+
# Translate Chinese text to English
|
| 27 |
+
translated_text = GoogleTranslator(source="zh-CN", target="en").translate(chinese_text)
|
| 28 |
+
|
| 29 |
+
# Convert the translated text to speech
|
| 30 |
+
tts = gTTS(text=translated_text, lang="en", slow=False)
|
| 31 |
+
tts.save("output.mp3")
|
| 32 |
+
|
| 33 |
+
return translated_text, "output.mp3"
|
| 34 |
+
|
| 35 |
+
# Create Gradio interfaces
|
| 36 |
+
interface1 = gr.Interface(
|
| 37 |
+
fn=recognize_and_translate,
|
| 38 |
+
inputs=gr.Audio(type="filepath"),
|
| 39 |
+
outputs="text",
|
| 40 |
+
title="Speech Recognition and Translation to Chinese",
|
| 41 |
+
)
|
| 42 |
+
|
| 43 |
+
interface2 = gr.Interface(
|
| 44 |
+
fn=translate_to_english,
|
| 45 |
+
inputs="text",
|
| 46 |
+
outputs=["text", "audio"],
|
| 47 |
+
title="Translate Chinese to English and Play",
|
| 48 |
+
)
|
| 49 |
+
|
| 50 |
+
# Combine both interfaces into a tabbed layout
|
| 51 |
+
app = gr.TabbedInterface([interface1, interface2], ["Speech to Chinese", "Chinese to English"])
|
| 52 |
+
|
| 53 |
+
# Launch the app
|
| 54 |
+
if __name__ == "__main__":
|
| 55 |
+
app.launch()
|