Spaces:
Running
Running
daavoo
commited on
Commit
•
0c2ec7e
1
Parent(s):
85f1ae3
Update app
Browse files
app.py
CHANGED
@@ -8,11 +8,11 @@ import streamlit as st
|
|
8 |
from document_to_podcast.preprocessing import DATA_LOADERS, DATA_CLEANERS
|
9 |
from document_to_podcast.inference.model_loaders import (
|
10 |
load_llama_cpp_model,
|
11 |
-
|
12 |
)
|
13 |
from document_to_podcast.config import DEFAULT_PROMPT, DEFAULT_SPEAKERS, Speaker
|
14 |
-
from document_to_podcast.inference.text_to_text import text_to_text_stream
|
15 |
from document_to_podcast.inference.text_to_speech import text_to_speech
|
|
|
16 |
|
17 |
|
18 |
@st.cache_resource
|
@@ -24,7 +24,7 @@ def load_text_to_text_model():
|
|
24 |
|
25 |
@st.cache_resource
|
26 |
def load_text_to_speech_model():
|
27 |
-
return
|
28 |
|
29 |
|
30 |
script = "script"
|
@@ -150,10 +150,10 @@ if uploaded_file is not None:
|
|
150 |
with st.spinner("Generating Audio..."):
|
151 |
speech = text_to_speech(
|
152 |
text.split(f'"Speaker {speaker_id}":')[-1],
|
153 |
-
|
154 |
-
voice_profile
|
155 |
)
|
156 |
-
st.audio(speech, sample_rate=speech_model.
|
157 |
|
158 |
st.session_state.audio.append(speech)
|
159 |
text = ""
|
@@ -164,7 +164,7 @@ if uploaded_file is not None:
|
|
164 |
sf.write(
|
165 |
"podcast.wav",
|
166 |
st.session_state.audio,
|
167 |
-
samplerate=speech_model.
|
168 |
)
|
169 |
st.markdown("Podcast saved to disk!")
|
170 |
|
|
|
8 |
from document_to_podcast.preprocessing import DATA_LOADERS, DATA_CLEANERS
|
9 |
from document_to_podcast.inference.model_loaders import (
|
10 |
load_llama_cpp_model,
|
11 |
+
load_outetts_model,
|
12 |
)
|
13 |
from document_to_podcast.config import DEFAULT_PROMPT, DEFAULT_SPEAKERS, Speaker
|
|
|
14 |
from document_to_podcast.inference.text_to_speech import text_to_speech
|
15 |
+
from document_to_podcast.inference.text_to_text import text_to_text_stream
|
16 |
|
17 |
|
18 |
@st.cache_resource
|
|
|
24 |
|
25 |
@st.cache_resource
|
26 |
def load_text_to_speech_model():
|
27 |
+
return load_outetts_model("OuteAI/OuteTTS-0.2-500M-GGUF/OuteTTS-0.2-500M-FP16.gguf")
|
28 |
|
29 |
|
30 |
script = "script"
|
|
|
150 |
with st.spinner("Generating Audio..."):
|
151 |
speech = text_to_speech(
|
152 |
text.split(f'"Speaker {speaker_id}":')[-1],
|
153 |
+
speech_model,
|
154 |
+
voice_profile,
|
155 |
)
|
156 |
+
st.audio(speech, sample_rate=speech_model.audio_codec.sr)
|
157 |
|
158 |
st.session_state.audio.append(speech)
|
159 |
text = ""
|
|
|
164 |
sf.write(
|
165 |
"podcast.wav",
|
166 |
st.session_state.audio,
|
167 |
+
samplerate=speech_model.audio_codec.sr,
|
168 |
)
|
169 |
st.markdown("Podcast saved to disk!")
|
170 |
|