Spaces:
Runtime error
Runtime error
File size: 2,563 Bytes
2720f57 bb7f34f 2720f57 dd56ef4 2720f57 ac45433 2720f57 3f76811 f60f6ff 550a372 3f76811 2720f57 b46bf98 2720f57 b46bf98 2720f57 e08bf83 3a682a4 07e0344 e08bf83 546b6d1 bb7f34f 11fb197 bb7f34f 5ea0c83 bb7f34f 2720f57 d72ad49 |
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 |
import gradio as gr
from deep_translator import GoogleTranslator
#import deepl
from transformers import pipeline
import os
from gtts import gTTS
from pytube import YouTube
pipe = pipeline(model="freeja/lab2-whisper-sv")
def transcribe_audio(audio,language):
transcribed = pipe(audio)["text"]
result = "Transcribed text\n"
result += transcribed + "\n"
result += "Translated text\n"
trans_text = translate_audio(transcribed,language)
result += trans_text
#text_to_speech(trans_text,language)
#result += text_to_speech
return result
def translate_audio(text,language):
#translate = deepl.Translator
language_dict = {"English":"en","Spanish":"es","German":"de","French":"fr","Italian":"it"}
lang = language_dict[language]
translated_text = GoogleTranslator(source='sv', target=lang).translate(text)
return translated_text
def text_to_speech(text,language):
language_dict = {"English":"en","Spanish":"es","German":"de","French":"fr","Italian":"it"}
lang = language_dict[language]
gTTS(text,lang,slow=False)
def transcribe_video(URL):
video = YouTube(str(URL))
yt = video.streams.filter(only_audio=True).first().download('youtube_video')
text = pipe(yt)["text"]
return text
"""iface = gr.Interface(
fn=transcribe_audio,
inputs=[
gr.Audio(source="microphone", type="filepath", label="Transcribe from Microphone"),
gr.Dropdown(["English","Spanish","Dutch","French","Italian"], value="English", label="Translate to ")
],
outputs="text",
title="Whisper Small Swedish",
description="Realtime demo for Swedish speech recognition with translation using a fine-tuned Whisper small model")"""
video_transcription = gr.Interface(
fn = transcribe_video,
inputs = "text",
outputs = "text",
title = "Whisper Small Swedish",
description = "Realtime demo for Swedish speech recognition with translation using a fine-tuned Whisper small model",
)
audio_transcription = gr.Interface(
fn=transcribe_audio,
inputs=[
gr.Audio(source="microphone", type="filepath", label="Transcribe from Microphone"),
gr.Dropdown(["English","Spanish","Dutch","French","Italian"], value="English", label="Translate to ")
],
outputs="text",
title="Whisper Small Swedish",
description="Realtime demo for Swedish speech recognition with translation using a fine-tuned Whisper small model",
)
iface = gr.TabbedInterface([audio_transcription, video_transcription], ["Transcribe Audio", "Transcribe Video"])
iface.launch() |