Spaces:
Sleeping
Sleeping
File size: 2,669 Bytes
fb652e4 | 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 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 | import streamlit as st
from st_audiorec import st_audiorec
# import whisper
import numpy as np
# model = whisper.load_model("base")
# def transcribe(audio):
# #time.sleep(3)
# # load audio and pad/trim it to fit 30 seconds
# audio = whisper.load_audio(audio)
# audio = whisper.pad_or_trim(audio)
# # make log-Mel spectrogram and move to the same device as the model
# mel = whisper.log_mel_spectrogram(audio).to(model.device)
# # detect the spoken language
# _, probs = model.detect_language(mel)
# print(f"Detected language: {max(probs, key=probs.get)}")
# # decode the audio
# options = whisper.DecodingOptions(fp16 = False)
# result = whisper.decode(model, mel, options)
# return result.text
import torch
from transformers import pipeline
# start model
device = "cuda:0" if torch.cuda.is_available() else "cpu"
translation_pipe = pipeline(
"automatic-speech-recognition", model="openai/whisper-base", device=device
)
def translate(audio, lang = "en"):
outputs = translation_pipe(audio, max_new_tokens=256, generate_kwargs={"task": "transcribe", "language": lang})
return outputs["text"]
# Customizing the Streamlit page layout
st.set_page_config(
page_title="tłumacz",
page_icon="🎤",
layout="wide",
)
# Center the app title and adjust padding
st.markdown(
"""
<style>
.stApp {
max-width: 1000px;
margin: 0 auto;
padding: 2rem;
}
.stTitle {
text-align: center;
}
</style>
""",
unsafe_allow_html=True
)
def audiorec_demo_app():
st.title('Audio Recorder App')
st.subheader('Record and Playback Audio')
# Add a little info on how to use the app
st.markdown(
"🎤 To record audio, simply click the 'Start Recording' button. "
"Once you're done, click 'Stop Recording' and 'Save Recording' to play it back. "
"You can also reset the recording if needed."
)
wav_audio_data = st_audiorec()
filename = "temp.wav"
if wav_audio_data is not None:
# Display the recorded audio
st.subheader('Transcribed text:')
with st.spinner("Transcribing..."):
with open(filename, "wb") as f:
f.write(wav_audio_data)
with open(filename, "rb") as f:
audio = f.read()
output_text = translate(audio, lang="pl")
translated_text = translate(audio, lang="en")
st.markdown(f"**Orginalny tekst [PL]:** {output_text}")
st.markdown(f"**Tłumaczenie [EN]:** {translated_text}")
if __name__ == '__main__':
audiorec_demo_app()
|