From 5feb9e0adcc64a4fabce9741c9cb5d100814846e Mon Sep 17 00:00:00 2001
From: insert <insert@insertapp.net>
Date: Mon, 12 Aug 2024 08:41:53 -0400
Subject: [PATCH] Update

---
 transcription.py | 27 ++++++++++++++-------------
 1 file changed, 14 insertions(+), 13 deletions(-)

diff --git a/transcription.py b/transcription.py
index ca005b6..fa59b07 100644
--- a/transcription.py
+++ b/transcription.py
@@ -6,6 +6,7 @@ import asyncio
 from pydub import AudioSegment
 import numpy as np
 import speech_recognition as sr
+import whisper
 
 def prepaudio(audiofile):
     st = sr.Recognizer()
@@ -21,21 +22,21 @@ class transcription(commands.Cog):
     def __init__(self, bot: commands.Bot):
         self.bot = bot
         self.st = sr.Recognizer()
+        self.model = whisper.load_model("base")
 
 
-    @nextcord.message_command(name="Transcribe Using Sphinx")
-    async def transcribesphinx(self, interaction: nextcord.Interaction, message: nextcord.Message):
-        try:
-            await interaction.response.defer(ephemeral='true')
-            await message.attachments[0].save("audio.ogg")
-            embed=nextcord.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 interaction.edit_original_message(embed=embed)
-            os.remove("audio.ogg")
-            os.remove("audio.wav")
-        except Exception as e:
-            await interaction.edit_original_message(content=f'an error appears to have occoured please report it to the developer: {e}')
+    @nextcord.message_command(name="Transcribe Using Whisper")
+    async def transcribewhisper(self, interaction: nextcord.Interaction, message: nextcord.Message):
+        await interaction.response.defer(ephemeral='true')
+        await message.attachments[0].save("audio.ogg")
+        oggfile = AudioSegment.from_ogg("audio.ogg")
+        oggfile.export("audio.mp3", format="mp3")
+        embed=nextcord.Embed(title="Audio Transcription",description=self.model.transcribe("audio.mp3")["text"], 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 interaction.edit_original_message(embed=embed)
+        os.remove("audio.ogg")
+        os.remove("audio.mp3")
 
     @nextcord.message_command(name="Transcribe Using Google")
     async def transcribegoogle(self, interaction: nextcord.Interaction, message: nextcord.Message):