57 lines
2.6 KiB
Python
57 lines
2.6 KiB
Python
|
import disnake
|
||
|
from disnake.ext import commands
|
||
|
import os
|
||
|
from os import path
|
||
|
import asyncio
|
||
|
from pydub import AudioSegment
|
||
|
import numpy as np
|
||
|
import speech_recognition as sr
|
||
|
|
||
|
def prepaudio(audiofile):
|
||
|
st = sr.Recognizer()
|
||
|
oggfile = AudioSegment.from_ogg(audiofile)
|
||
|
oggfile.export("audio.wav", format="wav")
|
||
|
convertemessage = sr.AudioFile("audio.wav")
|
||
|
with convertemessage as sounds:
|
||
|
transcribeaudo = st.record(sounds)
|
||
|
return transcribeaudo
|
||
|
|
||
|
class transcription(commands.Cog):
|
||
|
|
||
|
def __init__(self, bot: commands.Bot):
|
||
|
self.bot = bot
|
||
|
self.st = sr.Recognizer()
|
||
|
|
||
|
|
||
|
@commands.message_command(name="Transcribe Using Sphinx")
|
||
|
async def transcribesphinx(self, inter: disnake.ApplicationCommandInteraction, message: disnake.Message):
|
||
|
try:
|
||
|
await inter.response.defer(ephemeral='true')
|
||
|
await message.attachments[0].save("audio.ogg")
|
||
|
embed=disnake.Embed(title="Audio Transcription",description=self.st.recognize_sphinx(prepaudio("audio.ogg")), color=0x3584e4)
|
||
|
embed.set_author(name=message.author.display_name, url=message.jump_url, icon_url=message.author.display_avatar)
|
||
|
embed.set_footer(text="Accuracy not guaranteed")
|
||
|
await inter.edit_original_message(embed=embed)
|
||
|
os.remove("audio.ogg")
|
||
|
os.remove("audio.wav")
|
||
|
except Exception as e:
|
||
|
await inter.edit_original_message(content=f'an error appears to have occoured please report it to the developer: {e}')
|
||
|
|
||
|
@commands.message_command(name="Transcribe Using Google")
|
||
|
async def transcribegoogle(self, inter: disnake.ApplicationCommandInteraction, message: disnake.Message):
|
||
|
try:
|
||
|
await inter.response.defer(ephemeral='true')
|
||
|
await message.attachments[0].save("audio.ogg")
|
||
|
# WARNING Google is propritary, consider disabling however sphynix is currently not very good so this provides an option
|
||
|
embed=disnake.Embed(title="Audio Transcription",description=self.st.recognize_google(prepaudio("audio.ogg")), color=0x3584e4)
|
||
|
embed.set_author(name=message.author.display_name, url=message.jump_url, icon_url=message.author.display_avatar)
|
||
|
embed.set_footer(text="Accuracy not guaranteed")
|
||
|
await inter.edit_original_message(embed=embed)
|
||
|
await asyncio.sleep(3)
|
||
|
os.remove("audio.ogg")
|
||
|
os.remove("audio.wav")
|
||
|
except Exception as e:
|
||
|
await inter.edit_original_message(content=f'an error appears to have occoured please report it to the developer: {e}')
|
||
|
|
||
|
def setup(bot: commands.Bot):
|
||
|
bot.add_cog(transcription(bot))
|