Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
import torch
|
3 |
+
from transformers import pipeline, AutoTokenizer, AutoModelForSeq2SeqLM
|
4 |
+
import torchaudio
|
5 |
+
import os
|
6 |
+
|
7 |
+
def load_models():
|
8 |
+
st.session_state.transcription_pipe = pipeline(
|
9 |
+
task="automatic-speech-recognition",
|
10 |
+
model="alvanlii/whisper-small-cantonese",
|
11 |
+
chunk_length_s=60,
|
12 |
+
device="cuda" if torch.cuda.is_available() else "cpu"
|
13 |
+
)
|
14 |
+
st.session_state.transcription_pipe.model.config.forced_decoder_ids = st.session_state.transcription_pipe.tokenizer.get_decoder_prompt_ids(language="zh", task="transcribe")
|
15 |
+
|
16 |
+
st.session_state.translation_tokenizer = AutoTokenizer.from_pretrained("Helsinki-NLP/opus-mt-zh-en")
|
17 |
+
st.session_state.translation_model = AutoModelForSeq2SeqLM.from_pretrained("Helsinki-NLP/opus-mt-zh-en")
|
18 |
+
|
19 |
+
st.session_state.summary_pipe = pipeline("text-summarization", model="facebook/bart-large-cnn")
|
20 |
+
|
21 |
+
st.session_state.rating_pipe = pipeline("text-classification", model="cardiffnlp/twitter-roberta-base-sentiment-latest")
|
22 |
+
|
23 |
+
def transcribe_audio(audio_path):
|
24 |
+
pipe = st.session_state.transcription_pipe
|
25 |
+
return pipe(audio_path)["text"]
|
26 |
+
|
27 |
+
def translate_text(text):
|
28 |
+
tokenizer = st.session_state.translation_tokenizer
|
29 |
+
model = st.session_state.translation_model
|
30 |
+
inputs = tokenizer(text, return_tensors="pt")
|
31 |
+
outputs = model.generate(inputs["input_ids"], max_length=1000, num_beams=5)
|
32 |
+
return tokenizer.decode(outputs[0], skip_special_tokens=True)
|
33 |
+
|
34 |
+
def summarize_text(text):
|
35 |
+
return st.session_state.summary_pipe(text)[0]['summary_text']
|
36 |
+
|
37 |
+
def rate_quality(text):
|
38 |
+
result = st.session_state.rating_pipe(text)[0]
|
39 |
+
label_map = {"LABEL_0": "Poor", "LABEL_1": "Average", "LABEL_2": "Good"}
|
40 |
+
return label_map.get(result["label"], "Unknown")
|
41 |
+
|
42 |
+
def main():
|
43 |
+
st.title("Audio Processing & Conversation Quality Rating")
|
44 |
+
|
45 |
+
if "transcription_pipe" not in st.session_state:
|
46 |
+
with st.spinner("Loading models..."):
|
47 |
+
load_models()
|
48 |
+
|
49 |
+
uploaded_file = st.file_uploader("Upload an audio file", type=["wav", "mp3", "m4a"])
|
50 |
+
|
51 |
+
if uploaded_file is not None:
|
52 |
+
with st.spinner("Processing audio..."):
|
53 |
+
file_path = "temp_audio.wav"
|
54 |
+
with open(file_path, "wb") as f:
|
55 |
+
f.write(uploaded_file.read())
|
56 |
+
|
57 |
+
transcript = transcribe_audio(file_path)
|
58 |
+
translation = translate_text(transcript)
|
59 |
+
summary = summarize_text(translation)
|
60 |
+
rating = rate_quality(translation)
|
61 |
+
|
62 |
+
os.remove(file_path)
|
63 |
+
|
64 |
+
st.subheader("Transcription")
|
65 |
+
st.write(transcript)
|
66 |
+
|
67 |
+
st.subheader("Translation (English)")
|
68 |
+
st.write(translation)
|
69 |
+
|
70 |
+
st.subheader("Summary")
|
71 |
+
st.write(summary)
|
72 |
+
|
73 |
+
st.subheader("Conversation Quality Rating")
|
74 |
+
st.write(rating)
|
75 |
+
|
76 |
+
if __name__ == "__main__":
|
77 |
+
main()
|