Spaces:
Sleeping
Sleeping
File size: 2,644 Bytes
6dae90f aadeb1f 6dae90f aadeb1f 6dae90f aadeb1f 6dae90f 2433253 d1a4980 2433253 aadeb1f 2433253 aadeb1f 6dae90f 2433253 6dae90f 2433253 6dae90f 2433253 6dae90f 2433253 6dae90f 2433253 6dae90f 2433253 6dae90f 2433253 6dae90f aadeb1f 6dae90f |
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 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 |
from pathlib import Path
import numpy as np
import streamlit as st
from fastcore.xtras import globtastic
from annotator.utils import (
annotate,
df_from_result,
find_word_timestamp,
generate_srt,
get_audio,
get_v_from_url,
working_directory,
write_srt,
)
st.set_page_config(layout="wide")
SRT_PATH = Path("srt")
if not SRT_PATH.exists():
SRT_PATH.mkdir()
AUDIO_PATH = Path("./audio")
if not AUDIO_PATH.exists():
AUDIO_PATH.mkdir()
def make_sidebar():
with st.sidebar:
st.markdown("## yt-video-annotator")
st.write("Link to the GitHub repo")
def caption_from_url(url):
audio_src = get_audio(url)
v = get_v_from_url(url)
audio_src = globtastic(AUDIO_PATH, file_glob="*.wav", file_re=v)[0]
result = annotate(audio_src)
df = df_from_result(result)
return audio_src, df
def main():
url, name = None, None
make_sidebar()
place = "https://www.youtube.com/watch?v=C0DPdy98e4c&ab_channel=SimonYapp"
col1, col2 = st.columns([1.2, 1])
with col1:
url = st.text_input("Enter URL for the YT video", place)
st.video(url)
with col2:
default_opt = "Search for words"
opt = st.radio(
"What do you wish to do?",
[default_opt, "Generate subtitles for the entire video"],
)
if opt == default_opt:
st.markdown("### Search for words in the video")
words = st.text_input("Enter words separated by a comma")
words = words.split(",")
if st.button("Get Timestamps"):
audio_src, df = caption_from_url(url)
times = find_word_timestamp(df, *words)
times = np.asarray(times).reshape(len(words), -1)
# st.write(times)
for i, word in enumerate(words):
st.write(f"{word} is said on {times[i].flatten()} timestamp")
else:
if st.button("Generate SRT"):
audio_src, df = caption_from_url(url)
name = Path(audio_src).stem
s = generate_srt(df)
with working_directory(SRT_PATH):
write_srt(s, name)
if name is not None:
with working_directory(SRT_PATH):
key = get_v_from_url(url)
srt = globtastic(".", file_glob="*.srt", file_re=key)[0]
with open(srt) as f:
st.download_button("Download SRT", f, file_name=f"{name}.srt")
# subprocess.run(['rm', '-rf', 'audio'])
# subprocess.run(['rm', '-rf', 'srt'])
if __name__ == "__main__":
main()
|