DiamondYin's picture
Duplicate from Xhaheen/whisper-to-chatGPT
2a42563
raw
history blame
5.32 kB
from pyChatGPT import ChatGPT
import os
session_token = os.environ.get('SessionToken')
import whisper
whisper_model = whisper.load_model("small")
def chat_hf(audio, custom_token):
try:
whisper_text = translate(audio)
api = ChatGPT(session_token)
resp = api.send_message(whisper_text)
api.refresh_auth() # refresh the authorization token
api.reset_conversation() # reset the conversation
gpt_response = resp['message']
except:
whisper_text = translate(audio)
api = ChatGPT(custom_token)
resp = api.send_message(whisper_text)
api.refresh_auth() # refresh the authorization token
api.reset_conversation() # reset the conversation
gpt_response = resp['message']
return whisper_text, gpt_response
def translate(audio):
print("""
β€”
Sending audio to Whisper ...
β€”
""")
audio = whisper.load_audio(audio)
audio = whisper.pad_or_trim(audio)
mel = whisper.log_mel_spectrogram(audio).to(whisper_model.device)
_, probs = whisper_model.detect_language(mel)
transcript_options = whisper.DecodingOptions(task="transcribe", fp16 = False)
#translate_options = whisper.DecodingOptions(task="translate", fp16 = False)
transcription = whisper.decode(whisper_model, mel, transcript_options)
#translation = whisper.decode(whisper_model, mel, translate_options)
print("language spoken: " + transcription.language)
print("transcript: " + transcription.text)
print("β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”")
#print("translated: " + translation.text)
return transcription.text
title = """
<div style="text-align: center; max-width: 500px; margin: 0 auto;">
<div
style="
display: inline-flex;
align-items: center;
gap: 0.8rem;
font-size: 1.75rem;
margin-bottom: 10px;
"
>
<h1 style="font-weight: 600; margin-bottom: 7px;">
Whisper to chatGPT
</h1>
</div>
<p style="margin-bottom: 10px;font-size: 94%;font-weight: 100;line-height: 1.5em;">
Chat with GPT with your voice in your native language !
<br />If it fails enter custom session key see video for reference refer
<a href="https://youtu.be/TdNSj_qgdFk" target="_blank">Bhavesh Baht video</a>
</p>
<p style="font-size: 94%">
You can skip the queue by duplicating this space:
<span style="display: flex;align-items: center;justify-content: center;height: 30px;">
<a href="https://huggingface.co/fffiloni/whisper-to-chatGPT?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a>
</span>
</p>
</div>
"""
article = """
<div class="footer">
<p><a href="https://chat.openai.com/chat" target="_blank">chatGPT</a>
by <a href="https://openai.com/" style="text-decoration: underline;" target="_blank">OpenAI</a> -
Gradio Demo by πŸ€— <a href="https://twitter.com/fffiloni" target="_blank">Sylvain Filoni</a>
</p>
</div>
"""
css = '''
#col-container {max-width: 700px; margin-left: auto; margin-right: auto;}
a {text-decoration-line: underline; font-weight: 600;}
.footer {
margin-bottom: 45px;
margin-top: 35px;
text-align: center;
border-bottom: 1px solid #e5e5e5;
}
.footer>p {
font-size: .8rem;
display: inline-block;
padding: 0 10px;
transform: translateY(10px);
background: white;
}
.dark .footer {
border-color: #303030;
}
.dark .footer>p {
background: #0b0f19;
}
'''
import gradio as gr
with gr.Blocks(css=css) as demo:
with gr.Column(elem_id="col-container"):
gr.HTML(title)
with gr.Row():
record_input = gr.Audio(source="microphone",type="filepath", show_label=False)
send_btn = gr.Button("Send my message !")
custom_token = gr.Textbox(label='If it fails, use your own session token', placeholder="your own session token")
with gr.Column():
audio_translation = gr.Textbox(type="text",label="Whisper translation")
gpt_response = gr.Textbox(type="text",label="chatGPT response")
gr.HTML(article)
send_btn.click(chat_hf, inputs=[record_input, custom_token], outputs=[audio_translation, gpt_response])
demo.queue(max_size=32, concurrency_count=20).launch(debug=True)