Devin Xie commited on
Commit
d0b52af
1 Parent(s): a2f1abf

added sample inputs

Browse files
.gitignore CHANGED
@@ -1,2 +1 @@
1
- *.wav
2
  .voice_env
 
 
1
  .voice_env
app.py CHANGED
@@ -28,14 +28,21 @@ def main():
28
  title = f"""<h2 align="center" style="font-size: 1.2rem; margin-bottom: 2rem;">Make your favorite characters say anything!</h2>"""
29
  st.markdown(title, unsafe_allow_html=True)
30
 
 
 
 
 
 
 
31
  # Upload audio file
32
  uploaded_file = st.file_uploader('Add an audio file of the voice you want to clone...', type=['wav'])
 
 
 
33
 
34
- if uploaded_file is not None:
35
- reference_audio, synthesized_audio = st.columns(2)
36
- with reference_audio:
37
- st.header('Reference Audio')
38
- st.audio(uploaded_file, format='audio/wav')
39
 
40
  # Input text
41
  text_input = st.text_input('What do you want your character to say? (Alphabet letters only, DO NOT INCLUDE PUNCTUATION)')
@@ -44,11 +51,10 @@ def main():
44
  if text_input:
45
  try:
46
  with st.spinner('Synthesizing...'):
47
- output_path = generate_audio(uploaded_file, text_input)
48
 
49
- with synthesized_audio:
50
- st.header('Synthesized Audio')
51
- st.audio(output_path, format='audio/wav')
52
  except:
53
  st.error('There was an issue synthesizing the text. Please check the input and try again. Remember, do not include punctuation.')
54
  else:
 
28
  title = f"""<h2 align="center" style="font-size: 1.2rem; margin-bottom: 2rem;">Make your favorite characters say anything!</h2>"""
29
  st.markdown(title, unsafe_allow_html=True)
30
 
31
+ sample_files = {
32
+ 'Stewie Griffin': 'sample_inputs/stewie.wav',
33
+ 'Donald Trump': 'sample_inputs/trump.wav',
34
+ 'Joe Rogan': 'sample_inputs/rogan.wav'
35
+ }
36
+
37
  # Upload audio file
38
  uploaded_file = st.file_uploader('Add an audio file of the voice you want to clone...', type=['wav'])
39
+ selected_sample = st.selectbox('Or choose a sample:', list(sample_files.keys()))
40
+
41
+ speaker_file = sample_files[selected_sample] if uploaded_file is None else uploaded_file
42
 
43
+ if speaker_file:
44
+ st.header('Reference Audio')
45
+ st.audio(uploaded_file, format='audio/wav')
 
 
46
 
47
  # Input text
48
  text_input = st.text_input('What do you want your character to say? (Alphabet letters only, DO NOT INCLUDE PUNCTUATION)')
 
51
  if text_input:
52
  try:
53
  with st.spinner('Synthesizing...'):
54
+ output_path = generate_audio(speaker_file, text_input)
55
 
56
+ st.header('Synthesized Audio')
57
+ st.audio(output_path, format='audio/wav')
 
58
  except:
59
  st.error('There was an issue synthesizing the text. Please check the input and try again. Remember, do not include punctuation.')
60
  else:
sample_inputs/rogan.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7443710c741b5716007a7e89e0c9448779c23adbce258e78c84615dc0f7bd6d0
3
+ size 8559872
sample_inputs/stewie.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8559147ae1b5ac2b7925365e42604d05764fd812a214bdcc60d1169f67b380e0
3
+ size 5667310
sample_inputs/trump.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:235b7d7b4d40d0a80808644964a5f01dbca68ebc7043a61f03661cebe9061437
3
+ size 8698620