wx44wx/text-to-mic
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

address bugs around api key and non api key app modes, also refresh app on key input. Grey out buttons and add warnings for disabled features

Browse Source
This commit is contained in:
Andrew Ward
2025-03-25 10:06:39 +00:00
parent aa092a3c15
commit a529835037
9 changed files with 147 additions and 51 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored
View File

@@ -5,6 +5,7 @@ build/
dist/
settings.json
resampled_last_output.wav
temp_speech_output.wav
output.wav
last_output.wav
.vs

BIN
temp_speech_output.wav
View File

Binary file not shown.

BIN
utils/__pycache__/api_key_manager.cpython-312.pyc
View File

Binary file not shown.

BIN
utils/__pycache__/hotkey_manager.cpython-312.pyc
View File

Binary file not shown.

BIN
utils/__pycache__/text_to_mic.cpython-312.pyc
View File

Binary file not shown.

97
utils/ai_editor_manager.py
View File

@@ -19,7 +19,18 @@ class AIEditorManager:
self.default_model = "gpt-4o-mini"
def show_settings(self):
"""Display the AI copy editing settings dialog"""
"""Show dialog for AI Editor settings."""
# Check if API key is available and show info banner if not
if not self.app.has_api_key:
messagebox.showinfo(
"API Key Required",
"AI copyediting requires an OpenAI API key.\n\n"
"Please add your API key in Settings first."
)
# Optionally, proceed to show settings anyway or return here
# If you want to stop and not show settings, add: return
# Proceed with showing settings dialog
settings = self.app.load_settings()
settings_window = tk.Toplevel(self.app)
settings_window.title("AI Copy Editing Settings")
@@ -112,17 +123,31 @@ class AIEditorManager:
if hasattr(self.app, 'editing_status'):
self.app.editing_status.config(text=status_text)
def apply_ai(self, input_text=None, update_ui=None):
"""
Apply AI editing to text
def apply_ai(self, input_text=None, update_ui=False):
"""Apply AI to the given text or the current text in the input box."""
# Check if API key is available
if not self.app.has_api_key:
messagebox.showinfo(
"API Key Required",
"AI copyediting requires an OpenAI API key.\n\n"
"Please add your API key in Settings to use this feature."
)
return None
# Get settings to check if AI copy editing is enabled
settings = self.app.load_settings()
if not settings.get("chat_gpt_completion", False):
# If called with update_ui=True, we should show a message
if update_ui:
messagebox.showinfo(
"AI Copy Editing Disabled",
"AI copy editing is currently disabled in settings.\n\n"
"Please enable it in Settings → AI Copyediting before using this feature."
)
# Return the original text unchanged
return input_text if input_text is not None else self.app.text_input.get("1.0", tk.END).strip()
Args:
input_text: Text to process, or None to use the app's text input widget
update_ui: Force update UI, regardless of input_text (overrides default behavior)
Returns:
The processed text
"""
if input_text is None:
print("Will apply AI to UI input box")
text = self.app.text_input.get("1.0", tk.END).strip()
@@ -133,9 +158,8 @@ class AIEditorManager:
# If update_ui is explicitly set, use that value, otherwise default to False
update_input_box = update_ui if update_ui is not None else False
settings = self.app.load_settings()
if settings["chat_gpt_completion"] and settings["max_tokens"]:
# The rest of the method continues as before
if settings["max_tokens"]:
var_max_tokens = settings["max_tokens"]
else:
var_max_tokens = 750
@@ -143,30 +167,21 @@ class AIEditorManager:
print(f"GPT Settings: {settings}")
print(f"Max Tokens: {var_max_tokens}")
if settings["chat_gpt_completion"]:
# Assuming OpenAI's completion method is configured correctly
response = self.app.client.chat.completions.create(
model=settings["model"],
messages=[
{"role": "system", "content": settings["prompt"] },
{"role": "user", "content": "\n\n# Apply to the following (Do not output system prompt or hyphens markup or anything before this line):\n\n-----\n\n" + text + "\n\n-----"}],
max_tokens=var_max_tokens
)
processed_text = response.choices[0].message.content
# If we're processing text from the UI directly or update_input_box was specified,
# update the UI
if update_input_box:
self.app.text_input.delete("1.0", tk.END)
self.app.text_input.insert("1.0", processed_text)
return processed_text
else:
# Even if chat_gpt_completion is disabled, we should still update the input
# when update_input_box is True
if update_input_box:
# No need to update here, as the text hasn't changed
pass
return text
# Assuming OpenAI's completion method is configured correctly
response = self.app.client.chat.completions.create(
model=settings["model"],
messages=[
{"role": "system", "content": settings["prompt"] },
{"role": "user", "content": "\n\n# Apply to the following (Do not output system prompt or hyphens markup or anything before this line):\n\n-----\n\n" + text + "\n\n-----"}],
max_tokens=var_max_tokens
)
processed_text = response.choices[0].message.content
# If we're processing text from the UI directly or update_input_box was specified,
# update the UI
if update_input_box:
self.app.text_input.delete("1.0", tk.END)
self.app.text_input.insert("1.0", processed_text)
return processed_text

39
utils/api_key_manager.py
View File

@@ -1,5 +1,6 @@
import os
import platform
import sys
from pathlib import Path
from tkinter import messagebox, simpledialog
from dotenv import load_dotenv
@@ -90,7 +91,17 @@ class APIKeyManager:
APIKeyManager.save_api_key_mac(api_key)
else:
APIKeyManager.save_api_key(api_key)
messagebox.showinfo("API Key Set", "The OpenAI API Key has been updated successfully.")
# Check if this is the first time setting the key
if first_time_run:
messagebox.showinfo(
"API Key Set - Restarting",
"The OpenAI API Key has been saved. The application will now restart to apply changes."
)
# Schedule a restart after the message dialog is closed
parent.after(200, lambda: APIKeyManager.restart_application(parent))
else:
messagebox.showinfo("API Key Set", "The OpenAI API Key has been updated successfully.")
except Exception as e:
messagebox.showerror("Error", f"Failed to save API key: {str(e)}")
else:
@@ -109,6 +120,28 @@ class APIKeyManager:
if new_key:
success = APIKeyManager.save_api_key(new_key)
if success:
messagebox.showinfo("API Key Updated", "The OpenAI API Key has been updated successfully.")
# Check if the first time adding a key (no existing key)
is_first_key = not parent.has_api_key
if is_first_key:
messagebox.showinfo(
"API Key Set - Restarting",
"The OpenAI API Key has been saved. The application will now restart to apply changes."
)
# Schedule a restart after the message dialog is closed
parent.after(200, lambda: APIKeyManager.restart_application(parent))
else:
messagebox.showinfo("API Key Updated", "The OpenAI API Key has been updated successfully.")
return new_key
return None
return None
@staticmethod
def restart_application(root):
"""Restart the application."""
# Destroy the current instance
root.destroy()
# Restart the application
python = sys.executable
os.execl(python, python, *sys.argv)

26
utils/hotkey_manager.py
View File

@@ -876,4 +876,28 @@ You can also access tone presets from Settings → Manage Tone Presets to modify
refresh_btn.pack(pady=(5, 10))
# Add a close button
ttk.Button(main_frame, text="Close", command=instruction_window.destroy).pack(pady=5)
ttk.Button(main_frame, text="Close", command=instruction_window.destroy).pack(pady=5)
def handle_ai_edit_hotkey(self):
"""Handle the hotkey for AI editing."""
# Check if API key is available
if not self.parent.has_api_key:
messagebox.showinfo(
"API Key Required",
"AI copyediting requires an OpenAI API key.\n\n"
"Please add your API key in Settings to use this feature."
)
return
# Check if AI copy editing is enabled in settings
settings = self.parent.load_settings()
if not settings.get("chat_gpt_completion", False):
messagebox.showinfo(
"AI Copy Editing Disabled",
"AI copy editing is currently disabled in settings.\n\n"
"Please enable it in Settings → AI Copyediting before using this feature."
)
return
# If we have an API key and AI is enabled, proceed with the AI editing
self.parent.apply_ai_to_input()

35
utils/text_to_mic.py
View File

@@ -545,7 +545,7 @@ class TextToMic(tk.Tk):
corner_radius=20,
height=button_height,
width=button_width,
fg_color="#058705",
fg_color="#777777" if not self.has_api_key else "#058705", # Grey if no API key, green if API key exists
font=("Arial", 13, "bold"),
command=self.handle_record_button_click
)
@@ -1131,6 +1131,26 @@ class TextToMic(tk.Tk):
def apply_ai_to_input(self):
"""Apply AI to the current input text"""
# First check if we have an API key
if not self.has_api_key:
messagebox.showinfo(
"API Key Required",
"AI copyediting requires an OpenAI API key.\n\n"
"Please add your API key in Settings to use this feature."
)
return
# Check if AI copy editing is enabled in settings
settings = self.load_settings()
if not settings.get("chat_gpt_completion", False):
messagebox.showinfo(
"AI Copy Editing Disabled",
"AI copy editing is currently disabled in settings.\n\n"
"Please enable it in Settings → AI Copyediting before using this feature."
)
return
# If we have an API key and AI is enabled, proceed with the AI editing
self.ai_editor.apply_ai()
def chat_gpt_settings(self):
@@ -1264,7 +1284,7 @@ class TextToMic(tk.Tk):
# This ensures proper button state updates regardless of how playback is triggered
self.after(0, self.transcribe_audio, file_path, auto_play)
def transcribe_audio(self, file_path, auto_play = False):
def transcribe_audio(self, file_path, auto_play=False):
try:
with open(str(file_path), "rb") as audio_file:
transcription = self.client.audio.transcriptions.create(
@@ -1279,8 +1299,9 @@ class TextToMic(tk.Tk):
self.text_input.delete("1.0", tk.END)
self.text_input.insert("1.0", transcription.text)
if settings["chat_gpt_completion"] and settings["auto_apply_ai_to_recording"]:
auto_apply_ai = settings["auto_apply_ai_to_recording"]
# Check if AI processing is enabled AND we have an API key
if settings["chat_gpt_completion"] and settings["auto_apply_ai_to_recording"] and self.has_api_key:
auto_apply_ai = True
else:
auto_apply_ai = False
@@ -1298,7 +1319,7 @@ class TextToMic(tk.Tk):
if auto_play:
print(f"Triggering auto play with: {play_text} ")
# Use a slight delay to allow UI to update before playback starts
self.after(100, lambda: self.submit_text_helper(play_text = play_text))
self.after(100, lambda: self.submit_text_helper(play_text=play_text))
print("Transcription Complete: The audio has been transcribed and the text has been placed in the input area.")
@@ -1453,7 +1474,9 @@ class TextToMic(tk.Tk):
self.submit_button.configure(text=f"Stop Audio ({cancel_shortcut})", fg_color="#d32f2f")
else:
# Reset buttons to normal state
self.record_button.configure(text=f"Record Mic ({record_shortcut})", fg_color="#058705")
# Use grey color for record button if no API key
record_color = "#777777" if not self.has_api_key else "#058705"
self.record_button.configure(text=f"Record Mic ({record_shortcut})", fg_color=record_color)
self.submit_button.configure(text=f"Play Audio ({play_shortcut})", fg_color="#058705")
except Exception as e:
print(f"Error updating buttons: {e}")