import gradio as gr
from PIL import Image
import os
import openai
#api_key = os.environ.get('api_key')
from share_btn import community_icon_html, loading_icon_html, share_js
token = os.environ.get('HF_TOKEN')
whisper = gr.Interface.load(name="spaces/sanchit-gandhi/whisper-large-v2")
tts = gr.Interface.load(name="spaces/Flux9665/IMS-Toucan")
talking_face = gr.Blocks.load(name="spaces/fffiloni/one-shot-talking-face", api_key=token)
def infer(audio, openai_api_key):
whisper_result = whisper(audio, None, "translate", fn_index=0)
gpt_response = try_api(whisper_result, openai_api_key)
audio_response = tts(gpt_response[0], "English Text", "English Accent", "English Speaker's Voice", fn_index=0)
portrait_link = talking_face("wise_woman_portrait.png", audio_response, fn_index=0)
return gr.Textbox.update(value=whisper_result, visible=True), portrait_link, gr.Textbox.update(value=gpt_response[1], visible=True), gr.update(visible=True), gr.Button.update(visible=True)
def try_api(message, openai_api_key):
try:
response = call_api(message, openai_api_key)
return response, "no error"
except openai.error.Timeout as e:
#Handle timeout error, e.g. retry or log
print(f"OpenAI API request timed out: {e}")
return "oups", f"OpenAI API request timed out:
{e}"
except openai.error.APIError as e:
#Handle API error, e.g. retry or log
print(f"OpenAI API returned an API Error: {e}")
return "oups", f"OpenAI API returned an API Error:
{e}"
except openai.error.APIConnectionError as e:
#Handle connection error, e.g. check network or log
print(f"OpenAI API request failed to connect: {e}")
return "oups", f"OpenAI API request failed to connect:
{e}"
except openai.error.InvalidRequestError as e:
#Handle invalid request error, e.g. validate parameters or log
print(f"OpenAI API request was invalid: {e}")
return "oups", f"OpenAI API request was invalid:
{e}"
except openai.error.AuthenticationError as e:
#Handle authentication error, e.g. check credentials or log
print(f"OpenAI API request was not authorized: {e}")
return "oups", f"OpenAI API request was not authorized:
{e}"
except openai.error.PermissionError as e:
#Handle permission error, e.g. check scope or log
print(f"OpenAI API request was not permitted: {e}")
return "oups", f"OpenAI API request was not permitted:
{e}"
except openai.error.RateLimitError as e:
#Handle rate limit error, e.g. wait or log
print(f"OpenAI API request exceeded rate limit: {e}")
return "oups", f"OpenAI API request exceeded rate limit:
{e}"
def call_api(message, openai_api_key):
print("starting open ai")
openai.api_key = openai_api_key
response = openai.Completion.create(
model="text-davinci-003",
prompt=message,
temperature=0.5,
max_tokens=2048,
top_p=1,
frequency_penalty=0,
presence_penalty=0.6
)
return str(response.choices[0].text).split("\n",2)[2]
def clean_components():
return gr.Audio.update(value=None), gr.HTML.update(visible=False), gr.Textbox.update(visible=False), gr.Video.update(value=None), gr.Button.update(visible=False)
title = """
Use Whisper to ask, alive portrait responds !
Note: this demo is not able to sustain a conversation from earlier responses.
For more detailed results and dialogue, you should use the official ChatGPT interface.
—
Also, be aware that audio records from iOS devices will not be decoded as expected by Gradio. For the best experience, record your voice from a computer instead of your smartphone ;)