Update src/streamlit_app.py
Browse files- src/streamlit_app.py +18 -12
src/streamlit_app.py
CHANGED
@@ -3,16 +3,12 @@ os.environ["TRANSFORMERS_CACHE"] = "/app/.cache/huggingface"
|
|
3 |
os.environ["HF_HOME"] = "/app/.cache/huggingface"
|
4 |
os.environ["XDG_CACHE_HOME"] = "/app/.cache"
|
5 |
os.environ["XDG_CONFIG_HOME"] = "/app/.streamlit"
|
|
|
6 |
import torch
|
7 |
import torchaudio
|
8 |
from transformers import WhisperProcessor, WhisperForConditionalGeneration
|
9 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
10 |
import streamlit as st
|
11 |
-
import os
|
12 |
-
os.environ["TRANSFORMERS_CACHE"] = "/app/.cache/huggingface"
|
13 |
-
os.environ["HF_HOME"] = "/app/.cache/huggingface"
|
14 |
-
|
15 |
-
|
16 |
|
17 |
@st.cache_resource
|
18 |
def load_models():
|
@@ -36,20 +32,30 @@ def extract_text_features(text):
|
|
36 |
outputs = text_model(**inputs)
|
37 |
return outputs.logits.argmax(dim=1).item()
|
38 |
|
39 |
-
def predict_hate_speech(audio_path, text):
|
40 |
-
|
41 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
42 |
prediction = extract_text_features(text_input)
|
43 |
return "Hate Speech" if prediction == 1 else "Not Hate Speech"
|
44 |
|
45 |
st.title("Hate Speech Detector with Audio and Text")
|
46 |
-
audio_file = st.file_uploader("Upload an audio file", type=["wav", "mp3", "flac"])
|
47 |
text_input = st.text_input("Optional text input")
|
|
|
48 |
if st.button("Predict"):
|
49 |
if audio_file is not None:
|
50 |
-
with open("temp_audio
|
51 |
f.write(audio_file.read())
|
52 |
-
prediction = predict_hate_speech("temp_audio
|
|
|
|
|
|
|
53 |
st.success(prediction)
|
54 |
else:
|
55 |
-
st.warning("Please
|
|
|
3 |
os.environ["HF_HOME"] = "/app/.cache/huggingface"
|
4 |
os.environ["XDG_CACHE_HOME"] = "/app/.cache"
|
5 |
os.environ["XDG_CONFIG_HOME"] = "/app/.streamlit"
|
6 |
+
|
7 |
import torch
|
8 |
import torchaudio
|
9 |
from transformers import WhisperProcessor, WhisperForConditionalGeneration
|
10 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
11 |
import streamlit as st
|
|
|
|
|
|
|
|
|
|
|
12 |
|
13 |
@st.cache_resource
|
14 |
def load_models():
|
|
|
32 |
outputs = text_model(**inputs)
|
33 |
return outputs.logits.argmax(dim=1).item()
|
34 |
|
35 |
+
def predict_hate_speech(audio_path=None, text=None):
|
36 |
+
if text:
|
37 |
+
text_input = text
|
38 |
+
elif audio_path:
|
39 |
+
transcription = transcribe(audio_path)
|
40 |
+
text_input = transcription
|
41 |
+
else:
|
42 |
+
return "Please provide either audio or text input."
|
43 |
+
|
44 |
prediction = extract_text_features(text_input)
|
45 |
return "Hate Speech" if prediction == 1 else "Not Hate Speech"
|
46 |
|
47 |
st.title("Hate Speech Detector with Audio and Text")
|
48 |
+
audio_file = st.file_uploader("Upload an audio file (wav, mp3, flac, ogg, opus)", type=["wav", "mp3", "flac", "ogg", "opus"])
|
49 |
text_input = st.text_input("Optional text input")
|
50 |
+
|
51 |
if st.button("Predict"):
|
52 |
if audio_file is not None:
|
53 |
+
with open("temp_audio", "wb") as f:
|
54 |
f.write(audio_file.read())
|
55 |
+
prediction = predict_hate_speech("temp_audio", text_input)
|
56 |
+
st.success(prediction)
|
57 |
+
elif text_input:
|
58 |
+
prediction = predict_hate_speech(text=text_input)
|
59 |
st.success(prediction)
|
60 |
else:
|
61 |
+
st.warning("Please provide at least audio or text input.")
|