Spaces:
Runtime error
Runtime error
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import requests
|
2 |
+
import pytz
|
3 |
+
import streamlit as st
|
4 |
+
import os
|
5 |
+
|
6 |
+
from datetime import datetime
|
7 |
+
from audio_recorder_streamlit import audio_recorder
|
8 |
+
|
9 |
+
# Filepath for saving the text
|
10 |
+
file_path = '/mnt/data/text_output.txt'
|
11 |
+
|
12 |
+
API_URL = 'https://tonpixzfvq3791u9.us-east-1.aws.endpoints.huggingface.cloud'
|
13 |
+
headers = {
|
14 |
+
"Authorization": "Bearer XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
|
15 |
+
"Content-Type": "audio/wav"
|
16 |
+
}
|
17 |
+
|
18 |
+
def query(filename):
|
19 |
+
with open(filename, "rb") as f:
|
20 |
+
data = f.read()
|
21 |
+
response = requests.post(API_URL, headers=headers, data=data)
|
22 |
+
return response.json()
|
23 |
+
|
24 |
+
def generate_filename(prompt, file_type):
|
25 |
+
central = pytz.timezone('US/Central')
|
26 |
+
safe_date_time = datetime.now(central).strftime("%m%d_%H%M")
|
27 |
+
replaced_prompt = prompt.replace(" ", "_").replace("\n", "_")
|
28 |
+
safe_prompt = "".join(x for x in replaced_prompt if x.isalnum() or x == "_")[:90]
|
29 |
+
return f"{safe_date_time}_{safe_prompt}.{file_type}"
|
30 |
+
|
31 |
+
def save_and_play_audio(audio_recorder):
|
32 |
+
audio_bytes = audio_recorder()
|
33 |
+
if audio_bytes:
|
34 |
+
filename = generate_filename("Recording", "wav")
|
35 |
+
with open(filename, 'wb') as f:
|
36 |
+
f.write(audio_bytes)
|
37 |
+
st.audio(audio_bytes, format="audio/wav")
|
38 |
+
return filename
|
39 |
+
|
40 |
+
def transcribe_audio(filename):
|
41 |
+
output = query(filename)
|
42 |
+
return output
|
43 |
+
|
44 |
+
def save_transcription(transcription):
|
45 |
+
with open(file_path, 'a') as f:
|
46 |
+
f.write(f"{transcription}\n")
|
47 |
+
|
48 |
+
def load_previous_transcriptions():
|
49 |
+
if os.path.exists(file_path):
|
50 |
+
with open(file_path, 'r') as f:
|
51 |
+
return f.read()
|
52 |
+
return ""
|
53 |
+
|
54 |
+
def main():
|
55 |
+
st.title("Speech to Text π€π")
|
56 |
+
st.write("Record your speech and get the text. π¨οΈ")
|
57 |
+
|
58 |
+
previous_transcriptions = load_previous_transcriptions()
|
59 |
+
text_area = st.text_area("Transcriptions:", previous_transcriptions, height=400)
|
60 |
+
|
61 |
+
filename = save_and_play_audio(audio_recorder)
|
62 |
+
if filename is not None:
|
63 |
+
transcription = transcribe_audio(filename)
|
64 |
+
|
65 |
+
# Update the text area with new transcription
|
66 |
+
updated_transcriptions = f"{previous_transcriptions}\n{transcription}"
|
67 |
+
st.text_area("Transcriptions:", updated_transcriptions, height=400)
|
68 |
+
|
69 |
+
# Save the new transcription to file
|
70 |
+
save_transcription(transcription)
|
71 |
+
|
72 |
+
if __name__ == "__main__":
|
73 |
+
main()
|