cutted 0.2.1__py3-none-any.whl → 0.3.2__py3-none-any.whl

cutted/__init__.py

@@ -1,2 +1,2 @@
-__version__ = "0.1.0"
+__version__ = "0.3.2"
 __author__ = "simon0302010"

cutted/app.py

@@ -8,9 +8,19 @@ from matplotlib.backends.backend_tkagg import FigureCanvasTkAgg
 
 customtkinter.set_appearance_mode("Dark")
 
+try:
+    from .core import transcribe
+    whisper_support = True
+    print_info("Whisper support present.")
+except:
+    whisper_support = False
+    print_info("Whisper support is not present.")
+
 class CuttedApp:
     def __init__(self):
         self.AudioProcessor = audio_processor.AudioProcessor()
+        if whisper_support:
+            self.whisper = None
         self.gemini = gemini.GeminiClient()
         self.canvas = None
         self.cursor_line = None
@@ -44,12 +54,29 @@ class CuttedApp:
         
         undo_button = customtkinter.CTkButton(self.root, text="Undo", command=self.undo_last, width=70)
         undo_button.place(relx=0.1, rely=1.0, anchor="s", y=-30)
+        
+        if whisper_support:
+            self.use_transcript_checkbox = customtkinter.CTkCheckBox(
+                self.root,
+                text="Send transcript to Gemini (slower, more accurate)",
+                text_color="#888888",
+                font=("Arial", 12)
+            )
+            self.use_transcript_checkbox.place(relx=0.0, rely=1.0, anchor="w", y=-12)
+            
+        self.use_audio_checkbox = customtkinter.CTkCheckBox(
+            self.root,
+            text="Send audio to Gemini (buggy)",
+            text_color="#888888",
+            font=("Arial", 12)
+        )
+        self.use_audio_checkbox.place(relx=1.0, rely=1.0, anchor="e", y=-12)
 
         self.play_button = customtkinter.CTkButton(self.root, text="Play", command=self.play_audio, width=50)
         self.play_button.place(relx=0.3, rely=1.0, anchor="s", y=-30)
         self.stop_button = customtkinter.CTkButton(self.root, text="Stop", command=self.stop_audio, width=50)
         self.stop_button.place(relx=0.7, rely=1.0, anchor="s", y=-30)
-
+        
         self.input_frame = customtkinter.CTkFrame(self.root, fg_color="transparent", height=36)
         self.input_frame.place(relx=0.5, rely=1.0, anchor="s", y=-90, relwidth=0.8)
 
@@ -164,25 +191,36 @@ class CuttedApp:
             return
         
         text = self.entry.get()
-        full_prompt = f"You are a audio editing AI. You are controllable via natural language and editing a audio file. The audio file is {round(self.AudioProcessor.get_lenght())}s long."
-        full_prompt += f"\n\nUser Prompt: {text}"
-        self.entry.delete(0, "end")
-        
-        function_call, text_result = self.gemini.generate(full_prompt)
-        
-        if function_call:
-            print_info(f"Gemini called {function_call.name}")
-            if function_call.name == "cut_audio":
-                print_info("Cut function called")
-                args = function_call.args
-                result = self.AudioProcessor.cut(args["start"], args["end"])
-                if not result:
-                    messagebox.showerror("Error", "Please try again.")
-            self.update_plot()
-        elif text_result:
-            messagebox.showerror("Error", text_result.strip())
-        else:
-            print_fail("Gemini returned no data")
+        if text.strip():
+            full_prompt = f"You are a audio editing AI. You are controllable via natural language and editing a audio file. The audio file is {round(self.AudioProcessor.get_lenght())}s long."
+            if whisper_support:
+                if self.use_transcript_checkbox.get():
+                    if not self.whisper:
+                        messagebox.showinfo("Info", "Loading Whisper model. This may take a few minutes depending on your internet connection. See the progress in your command line. If this window appears to be frozen, the transcription is running.")
+                        self.whisper = transcribe.Whisper()
+                    transcript = self.whisper.transcribe(self.AudioProcessor.audio_path)
+                    full_prompt += f"\nThis is a transcript with per word timestamps of the audio:\n{transcript}"
+            full_prompt += f"\n\nUser Prompt: {text}"
+            self.entry.delete(0, "end")
+            
+            if self.use_audio_checkbox.get():
+                function_call, text_result = self.gemini.generate(full_prompt, audio_base64=self.AudioProcessor.get_audio_base64())
+            else:
+                function_call, text_result = self.gemini.generate(full_prompt)
+            
+            if function_call:
+                print_info(f"Gemini called {function_call.name}")
+                if function_call.name == "cut_audio":
+                    print_info("Cut function called")
+                    args = function_call.args
+                    result = self.AudioProcessor.cut(args["start"], args["end"])
+                    if not result:
+                        messagebox.showerror("Error", "Please try again.")
+                self.update_plot()
+            elif text_result:
+                messagebox.showerror("Error", text_result.strip())
+            else:
+                print_fail("Gemini returned no data")
 
     def save_state(self):
         if hasattr(self.AudioProcessor, "audio") and self.AudioProcessor.audio is not None:
@@ -205,4 +243,4 @@ class CuttedApp:
         
 def main():
     app = CuttedApp()
-    app.run()
+    app.run()

cutted/core/audio_processor.py

@@ -4,9 +4,8 @@ from .logger import *
 import numpy as np
 from matplotlib.figure import Figure
 import pygame
+import base64
 import io
-import threading
-import time
 
 class AudioProcessor:
     def __init__(self):
@@ -150,4 +149,12 @@ class AudioProcessor:
         }
         
     def export_audio(self, path, format: str = "mp3"):
-        self.audio.export(path, format=format)
+        self.audio.export(path, format=format)
+        
+    def get_audio_base64(self):
+        buffer = io.BytesIO()
+        self.audio.export(buffer, format="mp3")
+        buffer.seek(0)
+        audio_bytes = buffer.read()
+        audio_base64 = base64.b64encode(audio_bytes)
+        return audio_base64

cutted/core/gemini.py

@@ -1,5 +1,6 @@
 import os
 import sys
+import base64
 from google import genai
 from google.genai import types
 
@@ -15,13 +16,21 @@ class GeminiClient:
             api_key=GEMINI_API_KEY,
         )
     
-    def generate(self, prompt: str, model: str = "gemini-2.0-flash"):
+    def generate(self, prompt: str, model: str = "gemini-2.0-flash", audio_base64 = None):
+        parts=[
+            types.Part.from_text(text=prompt),
+        ]
+        
+        if audio_base64:
+            parts.append(types.Part.from_bytes(
+                mime_type="audio/mpeg",
+                data=base64.b64decode(audio_base64)
+            ))
+        
         contents = [
             types.Content(
                 role="user",
-                parts=[
-                    types.Part.from_text(text=prompt),
-                ],
+                parts=parts
             ),
         ]
         tools = [

cutted/core/transcribe.py

@@ -2,7 +2,7 @@ import json
 import whisper_timestamped as whisper
 
 class Whisper:
-    def __init__(self, model_size: str = "small", device: str = "cpu"):
+    def __init__(self, model_size: str = "small", device = None):
         self.model = whisper.load_model(
             model_size, device=device
         )

cutted-0.3.2.dist-info/METADATA

@@ -0,0 +1,55 @@
+Metadata-Version: 2.4
+Name: cutted
+Version: 0.3.2
+Summary: AI-powered audio editor controllable via natural language.
+Author-email: simon0302010 <simon0302010@gmail.com>
+License-Expression: GPL-3.0
+Project-URL: Homepage, https://github.com/simon0302010/Cutted
+Requires-Python: <=3.13,>=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: customtkinter
+Requires-Dist: matplotlib
+Requires-Dist: numpy
+Requires-Dist: pydub
+Requires-Dist: pygame
+Requires-Dist: google-genai
+Requires-Dist: python-dotenv
+Provides-Extra: whisper
+Requires-Dist: whisper-timestamped; extra == "whisper"
+Dynamic: license-file
+
+![PyPI](https://img.shields.io/pypi/v/cutted?color=blue)
+![PyPI - License](https://img.shields.io/pypi/l/lyriks-video)
+![Hackatime](https://hackatime-badge.hackclub.com/U08HC7N4JJW/Cutted)
+
+# Cutted
+
+AI-powered audio editor controlled by natural language 🚀
+
+Let AI handle your audio editing with simple commands:
+- Automatically detect quiet or loud parts
+- Transcribe audio (if Whisper is installed)
+- Cut, trim, or adjust volume for specific segments
+
+## Installation
+
+1. **Install system dependencies (e.g., FFmpeg).**  
+2. **Install Cutted**  
+```bash
+pip install cutted
+```
+**With Whisper support**  
+```bash
+pip install cutted[whisper]
+```
+  
+## Usage
+
+1. **Launch the app:**  
+   ```bash
+   python -m cutted
+   ```
+2. **Load an audio file** (MP3, WAV, etc.)  
+3. **Play, Cut, Undo** – all from the GUI or with text commands ✂️
+4. **Export** as MP3 or WAV  

cutted-0.3.2.dist-info/RECORD

@@ -0,0 +1,12 @@
+cutted/__init__.py,sha256=ttFLenYuOpXQTaR9nB0dF-3zFB_PeksXf9R4r_TB8S8,49
+cutted/__main__.py,sha256=lYGLgtIZ_vGZIJmWG6ZQoqOdyOJnaWEA4NBn5Rc7Q8E,61
+cutted/app.py,sha256=Ay_yVrPt1TQAE3lqmrII88lFoGZb5Mh5cDwediM9ZG8,9592
+cutted/core/audio_processor.py,sha256=JdPeWO_jAIn_uZFeZYQJX3RC0Vy8GClKrX7xGk4pXR4,5426
+cutted/core/gemini.py,sha256=yHsQXk4tDHcW5qJBuL2LpPvdChimxlmEbu76BvsoeY4,3108
+cutted/core/logger.py,sha256=AjqrgW2LV9HdPkPQ8oOmyd9lWzVSIg46r74ILR7mVHo,585
+cutted/core/transcribe.py,sha256=0e7aCva4y6D-gKe1xw5HT9VoFgbvHGgV6utn12r8wXA,986
+cutted-0.3.2.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+cutted-0.3.2.dist-info/METADATA,sha256=MPuSW4_LjNch0PXeFrGtRW1C4CuJ7unNv1ZZhvFHmSM,1503
+cutted-0.3.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+cutted-0.3.2.dist-info/top_level.txt,sha256=PL6glZvzRyKWCDn5aoYI9uH8HlEA5Qd_XFJowJKARYI,7
+cutted-0.3.2.dist-info/RECORD,,

cutted-0.2.1.dist-info/METADATA

@@ -1,66 +0,0 @@
-Metadata-Version: 2.4
-Name: cutted
-Version: 0.2.1
-Summary: AI-powered audio editor controllable via natural language.
-Author-email: simon0302010 <simon0302010@gmail.com>
-License-Expression: GPL-3.0
-Project-URL: Homepage, https://github.com/simon0302010/Cutted
-Requires-Python: <=3.13,>=3.9
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: customtkinter
-Requires-Dist: matplotlib
-Requires-Dist: numpy
-Requires-Dist: pydub
-Requires-Dist: pygame
-Requires-Dist: google-genai
-Requires-Dist: python-dotenv
-Dynamic: license-file
-
-# Cutted
-AI-powered audio editor controllable via natural language
-
-Note: This app is currently not functional.
-
-# Installation
-
-Install dependencies:
-
-**Debian:**
-```bash
-sudo apt update
-sudo apt install ffmpeg git
-```
-
-**Arch Linux:**
-```bash
-sudo pacman -Syu ffmpeg git
-```
-
-Clone the repository:
-
-```bash
-git clone https://github.com/simon0302010/Cutted.git
-cd Cutted
-```
-
-Create a virtual environment:
-
-```bash
-python -m venv venv
-source venv/bin/activate
-```
-
-Install the package:
-
-```bash
-pip install .
-```
-
-# Usage
-
-Run the package:
-
-```bash
-python -m cutted
-```

cutted-0.2.1.dist-info/RECORD

@@ -1,12 +0,0 @@
-cutted/__init__.py,sha256=CP0x3JIScNbFVSOoF3eIQTKD5gDRfWXcCFE46rlZCio,49
-cutted/__main__.py,sha256=lYGLgtIZ_vGZIJmWG6ZQoqOdyOJnaWEA4NBn5Rc7Q8E,61
-cutted/app.py,sha256=vJZ_HZtUffUw36tJFBJCAhBbdluQgaOWIPqM5dNZXgU,7706
-cutted/core/audio_processor.py,sha256=7-XCuPPTlozeuaD2LqyzwRGinu0NvowTLbAh2X4XJ98,5182
-cutted/core/gemini.py,sha256=Ts_EbC1-rO9jIsdSlzKcmjLVS1o663GmfTdzmix12kE,2872
-cutted/core/logger.py,sha256=AjqrgW2LV9HdPkPQ8oOmyd9lWzVSIg46r74ILR7mVHo,585
-cutted/core/transcribe.py,sha256=cm6ziM3_grXKpUCFHiAU7-6lFK_SVsf7-6n14vMYQng,992
-cutted-0.2.1.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-cutted-0.2.1.dist-info/METADATA,sha256=KKJ3auz3E5piy_1uEa9fWlkgZ72RmpHFMohbrqUmUbk,1122
-cutted-0.2.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-cutted-0.2.1.dist-info/top_level.txt,sha256=PL6glZvzRyKWCDn5aoYI9uH8HlEA5Qd_XFJowJKARYI,7
-cutted-0.2.1.dist-info/RECORD,,