pygpt-net 2.5.90__tar.gz → 2.5.91__tar.gz

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/CHANGELOG.md

@@ -1,5 +1,10 @@
 # CHANGELOG
 
+## 2.5.91 (2025-08-08)
+
+- Added GPT-5.
+- Added audio cache - #118.
+
 ## 2.5.90 (2025-08-07)
 
 - Fix: Initialize context summary if a conversation starts with a tool call.

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/PKG-INFO

@@ -1,9 +1,9 @@
 Metadata-Version: 2.3
 Name: pygpt-net
-Version: 2.5.90
-Summary: Desktop AI Assistant powered by: OpenAI o1, o3, GPT-4o, GPT-4 Vision, DALL-E 3, Llama 3, Mistral, Gemini, Claude, Grok, DeepSeek, Bielik, and other models supported by Llama Index, and Ollama. Chatbot, agents, completion, image generation, vision analysis, speech-to-text, plugins, internet access, file handling, command execution and more.
+Version: 2.5.91
+Summary: Desktop AI Assistant powered by: OpenAI GPT-5, o1, o3, GPT-4, Gemini, Claude, Grok, DeepSeek, and other models supported by Llama Index, and Ollama. Chatbot, agents, completion, image generation, vision analysis, speech-to-text, plugins, internet access, file handling, command execution and more.
 License: MIT
-Keywords: py_gpt,py-gpt,pygpt,desktop,app,o1,o3,gpt,gpt4,gpt-4o,gpt-4v,gpt3.5,gpt-4,gpt-4-vision,gpt-3.5,llama3,mistral,gemini,grok,deepseek,bielik,claude,tts,whisper,vision,chatgpt,dall-e,chat,chatbot,assistant,text completion,image generation,ai,api,openai,api key,langchain,llama-index,ollama,presets,ui,qt,pyside
+Keywords: py_gpt,py-gpt,pygpt,desktop,app,o1,o3,gpt-5,gpt,gpt4,gpt-4o,gpt-4v,gpt3.5,gpt-4,gpt-4-vision,gpt-3.5,llama3,mistral,gemini,grok,deepseek,bielik,claude,tts,whisper,vision,chatgpt,dall-e,chat,chatbot,assistant,text completion,image generation,ai,api,openai,api key,langchain,llama-index,ollama,presets,ui,qt,pyside
 Author: Marcin Szczyglinski
 Author-email: info@pygpt.net
 Requires-Python: >=3.10,<3.14
@@ -104,7 +104,7 @@ Description-Content-Type: text/markdown
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.5.90** | build: **2025-08-07** | Python: **>=3.10, <3.14**
+Release: **2.5.91** | build: **2025-08-08** | Python: **>=3.10, <3.14**
 
 > Official website: https://pygpt.net | Documentation: https://pygpt.readthedocs.io
 > 
@@ -116,7 +116,7 @@ Release: **2.5.90** | build: **2025-08-07** | Python: **>=3.10, <3.14**
 
 ## Overview
 
-**PyGPT** is **all-in-one** Desktop AI Assistant that provides direct interaction with OpenAI language models, including `o1`, `o3`, `gpt-4o`, `gpt-4`, `gpt-4 Vision`, and `gpt-3.5`, through the `OpenAI API`. By utilizing `LlamaIndex`, the application also supports alternative LLMs, like those available on `HuggingFace`, locally available models via `Ollama` (like `Llama 3`,`Mistral`, `DeepSeek V3/R1` or `Bielik`), `Google Gemini`, `Anthropic Claude`, `Perplexity Sonar`, and `xAI Grok`.
+**PyGPT** is **all-in-one** Desktop AI Assistant that provides direct interaction with OpenAI language models, including `gpt-5`, `o1`, `o3`, `gpt-4o`, `gpt-4`, `gpt-4 Vision`, and `gpt-3.5`, through the `OpenAI API`. By utilizing `LlamaIndex`, the application also supports alternative LLMs, like those available on `HuggingFace`, locally available models via `Ollama` (like `Llama 3`,`Mistral`, `DeepSeek V3/R1` or `Bielik`), `Google Gemini`, `Anthropic Claude`, `Perplexity Sonar`, and `xAI Grok`.
 
 This assistant offers multiple modes of operation such as chat, assistants, completions, and image-related tasks using `DALL-E 3` for generation and `gpt-4 Vision` for image analysis. **PyGPT** has filesystem capabilities for file I/O, can generate and run Python code, execute system commands, execute custom commands and manage file transfers. It also allows models to perform web searches with the `Google` and `Microsoft Bing`.
 
@@ -145,7 +145,7 @@ You can download compiled 64-bit versions for Windows and Linux here: https://py
 - Desktop AI Assistant for `Linux`, `Windows` and `Mac`, written in Python.
 - Works similarly to `ChatGPT`, but locally (on a desktop computer).
 - 12 modes of operation: Chat, Chat with Files, Chat with Audio, Research (Perplexity), Completion, Image generation, Vision, Assistants, Experts, Computer use, Agents and Autonomous Mode.
-- Supports multiple models: `o1`, `o3`, `GPT-4o`, `GPT-4`, `GPT-3.5`, and any model accessible through `LlamaIndex` and `Ollama` such as `Llama 3`, `Mistral`, `Google Gemini`, `Anthropic Claude`, `xAI Grok`, `DeepSeek V3/R1`, `Perplexity Sonar`, `Bielik`, etc.
+- Supports multiple models: `gpt-5`, `o1`, `o3`, `GPT-4o`, `GPT-4`, `GPT-3.5`, and any model accessible through `LlamaIndex` and `Ollama` such as `Llama 3`, `Mistral`, `Google Gemini`, `Anthropic Claude`, `xAI Grok`, `DeepSeek V3/R1`, `Perplexity Sonar`, `Bielik`, etc.
 - Chat with your own Files: integrated `LlamaIndex` support: chat with data such as: `txt`, `pdf`, `csv`, `html`, `md`, `docx`, `json`, `epub`, `xlsx`, `xml`, webpages, `Google`, `GitHub`, video/audio, images and other data types, or use conversation history as additional context provided to the model.
 - Built-in vector databases support and automated files and data embedding.
 - Included support features for individuals with disabilities: customizable keyboard shortcuts, voice control, and translation of on-screen actions into audio via speech synthesis.
@@ -1181,6 +1181,9 @@ PyGPT has built-in support for models (as of 2025-07-26):
 - `gpt-4o-2024-11-20` (OpenAI)
 - `gpt-4o-audio-preview` (OpenAI)
 - `gpt-4o-mini` (OpenAI)
+- `gpt-5` (OpenAI)
+- `gpt-5-mini` (OpenAI)
+- `gpt-5-nano` (OpenAI)
 - `gpt-image-1` (OpenAI)
 - `grok-2-vision` (xAI)
 - `grok-3` (xAI)
@@ -4351,6 +4354,11 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.5.91 (2025-08-08)**
+
+- Added GPT-5.
+- Added audio cache - #118.
+
 **2.5.90 (2025-08-07)**
 
 - Fix: Initialize context summary if a conversation starts with a tool call.

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/README.md

@@ -2,7 +2,7 @@
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.5.90** | build: **2025-08-07** | Python: **>=3.10, <3.14**
+Release: **2.5.91** | build: **2025-08-08** | Python: **>=3.10, <3.14**
 
 > Official website: https://pygpt.net | Documentation: https://pygpt.readthedocs.io
 > 
@@ -14,7 +14,7 @@ Release: **2.5.90** | build: **2025-08-07** | Python: **>=3.10, <3.14**
 
 ## Overview
 
-**PyGPT** is **all-in-one** Desktop AI Assistant that provides direct interaction with OpenAI language models, including `o1`, `o3`, `gpt-4o`, `gpt-4`, `gpt-4 Vision`, and `gpt-3.5`, through the `OpenAI API`. By utilizing `LlamaIndex`, the application also supports alternative LLMs, like those available on `HuggingFace`, locally available models via `Ollama` (like `Llama 3`,`Mistral`, `DeepSeek V3/R1` or `Bielik`), `Google Gemini`, `Anthropic Claude`, `Perplexity Sonar`, and `xAI Grok`.
+**PyGPT** is **all-in-one** Desktop AI Assistant that provides direct interaction with OpenAI language models, including `gpt-5`, `o1`, `o3`, `gpt-4o`, `gpt-4`, `gpt-4 Vision`, and `gpt-3.5`, through the `OpenAI API`. By utilizing `LlamaIndex`, the application also supports alternative LLMs, like those available on `HuggingFace`, locally available models via `Ollama` (like `Llama 3`,`Mistral`, `DeepSeek V3/R1` or `Bielik`), `Google Gemini`, `Anthropic Claude`, `Perplexity Sonar`, and `xAI Grok`.
 
 This assistant offers multiple modes of operation such as chat, assistants, completions, and image-related tasks using `DALL-E 3` for generation and `gpt-4 Vision` for image analysis. **PyGPT** has filesystem capabilities for file I/O, can generate and run Python code, execute system commands, execute custom commands and manage file transfers. It also allows models to perform web searches with the `Google` and `Microsoft Bing`.
 
@@ -43,7 +43,7 @@ You can download compiled 64-bit versions for Windows and Linux here: https://py
 - Desktop AI Assistant for `Linux`, `Windows` and `Mac`, written in Python.
 - Works similarly to `ChatGPT`, but locally (on a desktop computer).
 - 12 modes of operation: Chat, Chat with Files, Chat with Audio, Research (Perplexity), Completion, Image generation, Vision, Assistants, Experts, Computer use, Agents and Autonomous Mode.
-- Supports multiple models: `o1`, `o3`, `GPT-4o`, `GPT-4`, `GPT-3.5`, and any model accessible through `LlamaIndex` and `Ollama` such as `Llama 3`, `Mistral`, `Google Gemini`, `Anthropic Claude`, `xAI Grok`, `DeepSeek V3/R1`, `Perplexity Sonar`, `Bielik`, etc.
+- Supports multiple models: `gpt-5`, `o1`, `o3`, `GPT-4o`, `GPT-4`, `GPT-3.5`, and any model accessible through `LlamaIndex` and `Ollama` such as `Llama 3`, `Mistral`, `Google Gemini`, `Anthropic Claude`, `xAI Grok`, `DeepSeek V3/R1`, `Perplexity Sonar`, `Bielik`, etc.
 - Chat with your own Files: integrated `LlamaIndex` support: chat with data such as: `txt`, `pdf`, `csv`, `html`, `md`, `docx`, `json`, `epub`, `xlsx`, `xml`, webpages, `Google`, `GitHub`, video/audio, images and other data types, or use conversation history as additional context provided to the model.
 - Built-in vector databases support and automated files and data embedding.
 - Included support features for individuals with disabilities: customizable keyboard shortcuts, voice control, and translation of on-screen actions into audio via speech synthesis.
@@ -1079,6 +1079,9 @@ PyGPT has built-in support for models (as of 2025-07-26):
 - `gpt-4o-2024-11-20` (OpenAI)
 - `gpt-4o-audio-preview` (OpenAI)
 - `gpt-4o-mini` (OpenAI)
+- `gpt-5` (OpenAI)
+- `gpt-5-mini` (OpenAI)
+- `gpt-5-nano` (OpenAI)
 - `gpt-image-1` (OpenAI)
 - `grok-2-vision` (xAI)
 - `grok-3` (xAI)
@@ -4249,6 +4252,11 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.5.91 (2025-08-08)**
+
+- Added GPT-5.
+- Added audio cache - #118.
+
 **2.5.90 (2025-08-07)**
 
 - Fix: Initialize context summary if a conversation starts with a tool call.

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/pyproject.toml

@@ -1,14 +1,14 @@
 [tool.poetry]
 name = "pygpt-net"
-version = "2.5.90"
-description = "Desktop AI Assistant powered by: OpenAI o1, o3, GPT-4o, GPT-4 Vision, DALL-E 3, Llama 3, Mistral, Gemini, Claude, Grok, DeepSeek, Bielik, and other models supported by Llama Index, and Ollama. Chatbot, agents, completion, image generation, vision analysis, speech-to-text, plugins, internet access, file handling, command execution and more."
+version = "2.5.91"
+description = "Desktop AI Assistant powered by: OpenAI GPT-5, o1, o3, GPT-4, Gemini, Claude, Grok, DeepSeek, and other models supported by Llama Index, and Ollama. Chatbot, agents, completion, image generation, vision analysis, speech-to-text, plugins, internet access, file handling, command execution and more."
 authors = ["Marcin Szczyglinski <info@pygpt.net>"]
 license = "MIT"
 readme = "README.md"
 homepage = "https://pygpt.net"
 repository = "https://github.com/szczyglis-dev/py-gpt"
 documentation = "https://pygpt.readthedocs.io/"
-keywords = ["py_gpt", "py-gpt", "pygpt", "desktop", "app", "o1", "o3", "gpt", "gpt4", "gpt-4o", "gpt-4v", "gpt3.5", "gpt-4", "gpt-4-vision", "gpt-3.5", "llama3", "mistral", "gemini", "grok", "deepseek", "bielik", "claude", "tts", "whisper", "vision", "chatgpt", "dall-e", "chat",
+keywords = ["py_gpt", "py-gpt", "pygpt", "desktop", "app", "o1", "o3", "gpt-5", "gpt", "gpt4", "gpt-4o", "gpt-4v", "gpt3.5", "gpt-4", "gpt-4-vision", "gpt-3.5", "llama3", "mistral", "gemini", "grok", "deepseek", "bielik", "claude", "tts", "whisper", "vision", "chatgpt", "dall-e", "chat",
     "chatbot", "assistant", "text completion", "image generation", "ai", "api", "openai", "api key", "langchain", "llama-index", "ollama", "presets", "ui", "qt", "pyside"]
 classifiers = [
     "License :: OSI Approved :: MIT License",

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/src/pygpt_net/CHANGELOG.txt

@@ -1,3 +1,8 @@
+2.5.91 (2025-08-08)
+
+- Added GPT-5.
+- Added audio cache - #118.
+
 2.5.90 (2025-08-07)
 
 - Fix: Initialize context summary if a conversation starts with a tool call.

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/src/pygpt_net/__init__.py

@@ -6,15 +6,15 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.08.07 00:00:00                  #
+# Updated Date: 2025.08.08 00:00:00                  #
 # ================================================== #
 
 __author__ = "Marcin Szczygliński"
 __copyright__ = "Copyright 2025, Marcin Szczygliński"
 __credits__ = ["Marcin Szczygliński"]
 __license__ = "MIT"
-__version__ = "2.5.90"
-__build__ = "2025-08-07"
+__version__ = "2.5.91"
+__build__ = "2025-08-08"
 __maintainer__ = "Marcin Szczygliński"
 __github__ = "https://github.com/szczyglis-dev/py-gpt"
 __report__ = "https://github.com/szczyglis-dev/py-gpt/issues"

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/src/pygpt_net/controller/access/voice.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.08.07 03:00:00                  #
+# Updated Date: 2025.08.07 22:00:00                  #
 # ================================================== #
 
 from typing import Optional, List, Dict, Any
@@ -231,8 +231,7 @@ class Voice(QObject):
             self.switch_btn_stop()
 
             # stop audio output if playing
-            if self.window.controller.audio.is_playing():
-                self.window.controller.audio.stop_output()
+            self.window.controller.audio.stop_output()
 
             # set audio volume bar
             self.window.core.audio.capture.set_bar(

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/src/pygpt_net/controller/audio/audio.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.01.17 13:00:00                  #
+# Updated Date: 2025.08.07 22:00:00                  #
 # ================================================== #
 
 import os
@@ -291,7 +291,7 @@ class Audio:
 
         if use_cache:
             lang = self.window.core.config.get("lang")
-            cache_dir = os.path.join(self.window.core.config.get_user_path(), "cache", "audio", lang)
+            cache_dir = os.path.join(self.window.core.audio.get_cache_dir(), lang)
             if not os.path.exists(cache_dir):
                 os.makedirs(cache_dir, exist_ok=True)
             cache_file = os.path.join(str(cache_dir), event.name + ".wav")
@@ -318,7 +318,7 @@ class Audio:
             )
             return
 
-        cache_dir = os.path.join(self.window.core.config.get_user_path(), "cache", "audio")
+        cache_dir = self.window.core.audio.get_cache_dir()
         if os.path.exists(cache_dir):
             import shutil
             shutil.rmtree(cache_dir)
@@ -352,7 +352,10 @@ class Audio:
 
         :param text: text to play
         """
-        self.window.update_status(trans("status.audio.start"))
+        if text:
+            self.window.update_status(trans("status.audio.start"))
+        else:
+            self.window.update_status("")
         QApplication.processEvents()  # process events to update UI
 
     def on_play(self, event: str):

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/src/pygpt_net/controller/ctx/extra.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.07.14 18:00:00                  #
+# Updated Date: 2025.08.07 22:00:00                  #
 # ================================================== #
 
 from PySide6.QtWidgets import QApplication
@@ -220,12 +220,7 @@ class Extra:
                 self.window.controller.audio.read_text(item.output)
                 self.audio_play_id = item_id
             elif self.audio_play_id == item_id:
-                if not self.window.controller.audio.is_playing():
-                    self.window.controller.audio.read_text(item.output)
-                else:
-                    self.window.controller.audio.stop_output()
-                    self.window.controller.audio.on_stop()
-                    self.audio_play_id = None
+                self.window.controller.audio.read_text(item.output)
 
     def join_item(
             self,

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/src/pygpt_net/controller/kernel/kernel.py

@@ -52,6 +52,7 @@ class Kernel(QObject):
             KernelEvent.APPEND_DATA, 
             KernelEvent.INPUT_USER,
             KernelEvent.FORCE_CALL,
+            KernelEvent.STATUS,
         ]
 
     def init(self):

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/src/pygpt_net/core/audio/audio.py

@@ -6,9 +6,11 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.08.07 03:00:00                  #
+# Updated Date: 2025.08.07 22:00:00                  #
 # ================================================== #
 
+import hashlib
+import os
 import re
 from typing import Union, Optional, Tuple, List
 
@@ -21,6 +23,9 @@ from .whisper import Whisper
 
 
 class Audio:
+
+    CACHE_FORMAT = "mp3"  # default cache format
+
     def __init__(self, window=None):
         """
         Audio input/output core
@@ -173,3 +178,63 @@ class Audio:
         :return: Error
         """
         return self.last_error
+
+    def prepare_cache_path(self, content: str) -> Tuple[Union[str, None], bool]:
+        """
+        Prepare unique cache file name for given content
+
+        :param content: text content to generate cache file name
+        :return: cache file path or None if content is empty
+        """
+        exists = False
+        if not content:
+            return None, exists
+        sha1sum = hashlib.sha1(content.encode('utf-8')).hexdigest()
+        filename = f"{sha1sum}." + self.CACHE_FORMAT
+        tmp_dir = self.get_cache_dir()
+        path = os.path.join(tmp_dir, filename)
+        if os.path.exists(path):
+            exists = True
+        return str(path), exists
+
+    def get_cache_dir(self) -> str:
+        """
+        Get cache directory for audio files
+
+        :return: audio cache directory path
+        """
+        dir = self.window.core.config.get_user_dir("tmp")
+        tmp_dir = os.path.join(dir, "audio_cache")
+        if not os.path.exists(tmp_dir):
+            os.makedirs(tmp_dir, exist_ok=True)
+        return tmp_dir
+
+    def mp3_to_wav(
+            self,
+            src_file: str,
+            dst_file: Optional[str] = None
+    ) -> Union[str, None]:
+        """
+        Convert MP3 file to WAV format
+
+        :param src_file: Path to the source MP3 file
+        :param dst_file: Optional path for the destination WAV file.
+        :return: Path to the converted WAV file or None if conversion fails.
+        """
+        from pydub import AudioSegment
+        try:
+            mp3_audio = AudioSegment.from_mp3(src_file)
+        except Exception as e:
+            print(f"Error loading mp3 file: {e}")
+            print("Please install ffmpeg to handle mp3 files: https://ffmpeg.org/")
+            return
+        if dst_file is None:
+            dir = os.path.dirname(src_file)
+            filename = os.path.splitext(os.path.basename(src_file))[0] + ".wav"
+            dst_file = os.path.join(dir, filename)
+        try:
+            mp3_audio.export(dst_file, format="wav")
+            return str(dst_file)
+        except Exception as e:
+            print(f"Error exporting wav file: {e}")
+            return

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/src/pygpt_net/core/audio/backend/native.py

@@ -6,18 +6,21 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.08.07 03:00:00                  #
+# Updated Date: 2025.08.07 22:00:00                  #
 # ================================================== #
 
+import os
 import time
 from typing import List, Tuple
 
 from PySide6.QtCore import QTimer, QObject
+from pydub import AudioSegment
 
 
 class NativeBackend(QObject):
 
     MIN_FRAMES = 25  # minimum frames to start transcription
+    AUTO_CONVERT_TO_WAV = False  # automatically convert to WAV format
 
     def __init__(self, window=None):
         """
@@ -42,7 +45,10 @@ class NativeBackend(QObject):
         self.stop_callback = None
         self.player = None
         self.playback_timer = None
+        self.volume_timer = None
         self.audio_output = None
+        self.envelope = []
+        self.chunk_ms = 10
 
         # Get configuration values (use defaults if unavailable)
         if self.window is not None and hasattr(self.window, "core"):
@@ -457,8 +463,6 @@ class NativeBackend(QObject):
         :param signals: Signals to emit on playback
         :return: True if started
         """
-        import os
-        from pydub import AudioSegment
         from PySide6.QtMultimedia import QMediaPlayer, QAudioOutput, QMediaDevices
         from PySide6.QtCore import QUrl, QTimer
         if signals is not None:
@@ -476,42 +480,57 @@ class NativeBackend(QObject):
             selected_device = devices[num_device] if num_device < len(devices) else devices[0]
             self.audio_output.setDevice(selected_device)
 
-        self.player = QMediaPlayer()
-        self.player.setAudioOutput(self.audio_output)
-
-        if audio_file.lower().endswith('.mp3'):
-            try:
-                mp3_audio = AudioSegment.from_mp3(audio_file)
-            except Exception as e:
-                print(f"Error loading mp3 file: {e}")
-                return
-
-            base_dir = self.window.core.config.get_user_path()
-            base_name = os.path.splitext(os.path.basename(audio_file))[0]
-            wav_file = os.path.join(base_dir, "_" + base_name + ".wav")
-            try:
-                mp3_audio.export(wav_file, format="wav")
-            except Exception as e:
-                print(f"Error exporting wav file: {e}")
-                return
-
-            self.player.setSource(QUrl.fromLocalFile(wav_file))
-        else:
-            self.player.setSource(QUrl.fromLocalFile(audio_file))
-
-        self.player.play()
-        self.playback_timer = QTimer()
-        self.playback_timer.setInterval(100)
+        if self.AUTO_CONVERT_TO_WAV:
+            if audio_file.lower().endswith('.mp3'):
+                tmp_dir = self.window.core.audio.get_cache_dir()
+                base_name = os.path.splitext(os.path.basename(audio_file))[0]
+                dst_file = os.path.join(tmp_dir, "_" + base_name + ".wav")
+                wav_file = self.window.core.audio.mp3_to_wav(audio_file, dst_file)
+                if wav_file:
+                    audio_file = wav_file
 
         def check_stop():
             if stopped():
                 self.player.stop()
-                self.playback_timer.stop()
+                self.stop_timers()
+                signals.volume_changed.emit(0)
+            else:
+                if self.player:
+                    if self.player.playbackState() == QMediaPlayer.StoppedState:
+                        self.player.stop()
+                        self.stop_timers()
+                        signals.volume_changed.emit(0)
 
+        self.envelope = self.calculate_envelope(audio_file, self.chunk_ms)
+        self.player = QMediaPlayer()
+        self.player.setAudioOutput(self.audio_output)
+        self.player.setSource(QUrl.fromLocalFile(audio_file))
+        self.player.play()
+
+        self.playback_timer = QTimer()
+        self.playback_timer.setInterval(100)
         self.playback_timer.timeout.connect(check_stop)
+        self.volume_timer = QTimer(self)
+        self.volume_timer.setInterval(10)  # every 100 ms
+        self.volume_timer.timeout.connect(
+            lambda: self.update_volume(signals)
+        )
+
         self.playback_timer.start()
+        self.volume_timer.start()
         signals.volume_changed.emit(0)
 
+    def stop_timers(self):
+        """
+        Stop playback timers.
+        """
+        if self.playback_timer is not None:
+            self.playback_timer.stop()
+            self.playback_timer = None
+        if self.volume_timer is not None:
+            self.volume_timer.stop()
+            self.volume_timer = None
+
     def play(
             self,
             audio_file: str,
@@ -543,6 +562,49 @@ class NativeBackend(QObject):
                 self.playback_timer.stop()
         return False
 
+    def calculate_envelope(
+            self,
+            audio_file: str,
+            chunk_ms: int = 100
+    ) -> list:
+        """
+        Calculate the volume envelope of an audio file.
+
+        :param audio_file: Path to the audio file
+        :param chunk_ms: Size of each chunk in milliseconds
+        """
+        import numpy as np
+        audio = AudioSegment.from_file(audio_file)
+        max_amplitude = 32767
+        envelope = []
+
+        for ms in range(0, len(audio), chunk_ms):
+            chunk = audio[ms:ms + chunk_ms]
+            rms = chunk.rms
+            if rms > 0:
+                db = 20 * np.log10(rms / max_amplitude)
+            else:
+                db = -60
+            db = max(-60, min(0, db))
+            volume = ((db + 60) / 60) * 100
+            envelope.append(volume)
+
+        return envelope
+
+    def update_volume(self, signals=None):
+        """
+        Update the volume based on the current position in the audio file.
+
+        :param signals: Signals object to emit volume changed event.
+        """
+        pos = self.player.position()
+        index = int(pos / self.chunk_ms)
+        if index < len(self.envelope):
+            volume = self.envelope[index]
+        else:
+            volume = 0
+        signals.volume_changed.emit(volume)
+
     def get_input_devices(self) -> List[Tuple[int, str]]:
         """
         Get input devices

{pygpt_net-2.5.90 → pygpt_net-2.5.91}/src/pygpt_net/data/config/config.json

@@ -1,8 +1,8 @@
 {
   "__meta__": {
-    "version": "2.5.90",
-    "app.version": "2.5.90",
-    "updated_at": "2025-08-07T00:00:00"
+    "version": "2.5.91",
+    "app.version": "2.5.91",
+    "updated_at": "2025-08-08T00:00:00"
   },
   "access.audio.event.speech": false,
   "access.audio.event.speech.disabled": [],
@@ -94,6 +94,7 @@
   "attachments_auto_index": true,
   "attachments_send_clear": true,
   "attachments_capture_clear": true,
+  "audio.cache.enabled": true,
   "audio.input.backend": "native",
   "audio.input.channels": 1,
   "audio.input.continuous": false,