Spaces:
Sleeping
Sleeping
File size: 2,016 Bytes
d73b234 a105247 d73b234 a105247 d73b234 a105247 d73b234 a105247 d73b234 a105247 d73b234 a105247 d73b234 11d063e a105247 11d063e d73b234 a105247 d73b234 169b7bd d73b234 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 |
# import streamlit as st
# import whisper
# from tempfile import NamedTemporaryFile
# import ffmpeg
# st.title("MinuteBot App")
# # upload audio file with streamlit
# audio_file = st.file_uploader("Unggah Meeting Audio", type=["mp3", "wav", "m4a"])
# # model = whisper.load_model("base") # loading the base model
# st.text("MinuteBot Model telah dimuat:")
# def load_whisper_model():
# return model
# if st.sidebar.button("Transkripsikan Audio"):
# if audio_file is not None:
# with NamedTemporaryFile() as temp:
# temp.write(audio_file.getvalue())
# temp.seek(0)
# model = whisper.load_model("large")
# result = model.transcribe(temp.name)
# st.write(result["text"])
# st.sidebar.header("Putar Berkas Audio")
# st.sidebar.audio(audio_file)
import streamlit as st
from tempfile import NamedTemporaryFile
import ffmpeg
from transformers import Wav2Vec2ForCTC, Wav2Vec2Tokenizer
import librosa
st.title("TemplarX-Medium-Indonesian Transcription App")
st.text("Model Whisper (TemplarX-medium-Indonesian) telah dimuat:")
def load_whisper_model():
model_name = "jonnatakusuma/TemplarX-medium-Indonesian"
tokenizer = Wav2Vec2Tokenizer.from_pretrained(model_name)
model = Wav2Vec2ForCTC.from_pretrained(model_name)
return tokenizer, model
audio_file = st.file_uploader("Unggah Meeting Audio", type=["mp3", "wav", "m4a"])
if st.sidebar.button("Transkripsikan Audio"):
if audio_file is not None:
with NamedTemporaryFile() as temp:
temp.write(audio_file.read())
temp.seek(0)
tokenizer, model = load_whisper_model()
# Read the audio file and transcribe using the fine-tuned model
audio_path = temp.name
audio_input, _ = librosa.load(audio_path, sr=16000)
transcription = model.stt(text)
st.write(transcription)
st.sidebar.header("Putar Berkas Audio")
st.sidebar.audio(audio_file, format='audio/wav')
|