pygpt-net 2.4.52__tar.gz → 2.4.54__tar.gz

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/CHANGELOG.md

@@ -1,5 +1,14 @@
 # CHANGELOG
 
+## 2.4.54 (2025-01-18)
+
+- Audio output switched from PyGame to PyAudio. It may be necessary to manually connect Alsa in Snap version with: "sudo snap connect pygpt:alsa".
+- Added audio output volume progress bar.
+
+## 2.4.53 (2025-01-17)
+
+- Fix: issue #89
+
 ## 2.4.52 (2025-01-17)
 
 - Improved audio input button visibility toggle.

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pygpt-net
-Version: 2.4.52
+Version: 2.4.54
 Summary: Desktop AI Assistant powered by models: OpenAI o1, GPT-4o, GPT-4, GPT-4 Vision, GPT-3.5, DALL-E 3, Llama 3, Mistral, Gemini, Claude, Bielik, and other models supported by Langchain, Llama Index, and Ollama. Features include chatbot, text completion, image generation, vision analysis, speech-to-text, internet access, file handling, command execution and more.
 Home-page: https://pygpt.net
 License: MIT
@@ -93,7 +93,7 @@ Description-Content-Type: text/markdown
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.4.52** | build: **2025.01.17** | Python: **>=3.10, <3.13**
+Release: **2.4.54** | build: **2025.01.18** | Python: **>=3.10, <3.13**
 
 > Official website: https://pygpt.net | Documentation: https://pygpt.readthedocs.io
 > 
@@ -212,6 +212,13 @@ sudo snap connect pygpt:audio-record :audio-record
 sudo snap connect pygpt:alsa
 ```
 
+**Using audio output:** to use audio output in Snap version you must connect the audio with:
+
+```commandline
+sudo snap connect pygpt:audio-playback
+sudo snap connect pygpt:alsa
+```
+
 **Connecting IPython in Docker in Snap version**:
 
 To use IPython in the Snap version, you must connect PyGPT to the Docker daemon:
@@ -4044,6 +4051,15 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.4.54 (2025-01-18)**
+
+- Audio output switched from PyGame to PyAudio. It may be necessary to manually connect Alsa in Snap version with: "sudo snap connect pygpt:alsa".
+- Added audio output volume progress bar.
+
+**2.4.53 (2025-01-17)**
+
+- Fix: issue #89
+
 **2.4.52 (2025-01-17)**
 
 - Improved audio input button visibility toggle.

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/README.md

@@ -2,7 +2,7 @@
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.4.52** | build: **2025.01.17** | Python: **>=3.10, <3.13**
+Release: **2.4.54** | build: **2025.01.18** | Python: **>=3.10, <3.13**
 
 > Official website: https://pygpt.net | Documentation: https://pygpt.readthedocs.io
 > 
@@ -121,6 +121,13 @@ sudo snap connect pygpt:audio-record :audio-record
 sudo snap connect pygpt:alsa
 ```
 
+**Using audio output:** to use audio output in Snap version you must connect the audio with:
+
+```commandline
+sudo snap connect pygpt:audio-playback
+sudo snap connect pygpt:alsa
+```
+
 **Connecting IPython in Docker in Snap version**:
 
 To use IPython in the Snap version, you must connect PyGPT to the Docker daemon:
@@ -3953,6 +3960,15 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.4.54 (2025-01-18)**
+
+- Audio output switched from PyGame to PyAudio. It may be necessary to manually connect Alsa in Snap version with: "sudo snap connect pygpt:alsa".
+- Added audio output volume progress bar.
+
+**2.4.53 (2025-01-17)**
+
+- Fix: issue #89
+
 **2.4.52 (2025-01-17)**
 
 - Improved audio input button visibility toggle.

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pygpt-net"
-version = "2.4.52"
+version = "2.4.54"
 description = "Desktop AI Assistant powered by models: OpenAI o1, GPT-4o, GPT-4, GPT-4 Vision, GPT-3.5, DALL-E 3, Llama 3, Mistral, Gemini, Claude, Bielik, and other models supported by Langchain, Llama Index, and Ollama. Features include chatbot, text completion, image generation, vision analysis, speech-to-text, internet access, file handling, command execution and more."
 authors = ["Marcin Szczyglinski <info@pygpt.net>"]
 license = "MIT"

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/CHANGELOG.txt

@@ -1,3 +1,12 @@
+2.4.54 (2025-01-18)
+
+- Audio output switched from PyGame to PyAudio. It may be necessary to manually connect Alsa in Snap version with: "sudo snap connect pygpt:alsa".
+- Added audio output volume progress bar.
+
+2.4.53 (2025-01-17)
+
+- Fix: issue #89
+
 2.4.52 (2025-01-17)
 
 - Improved audio input button visibility toggle.

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/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: 2025.01.17 13:00:00                  #
+# Updated Date: 2025.01.18 03:00:00                  #
 # ================================================== #
 
 __author__ = "Marcin Szczygliński"
 __copyright__ = "Copyright 2025, Marcin Szczygliński"
 __credits__ = ["Marcin Szczygliński"]
 __license__ = "MIT"
-__version__ = "2.4.52"
-__build__ = "2025.01.17"
+__version__ = "2.4.54"
+__build__ = "2025.01.18"
 __maintainer__ = "Marcin Szczygliński"
 __github__ = "https://github.com/szczyglis-dev/py-gpt"
 __report__ = "https://github.com/szczyglis-dev/py-gpt/issues"

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/controller/dialogs/info.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.01.16 01:00:00                  #
+# Updated Date: 2025.01.17 13:00:00                  #
 # ================================================== #
 
 import webbrowser
@@ -113,7 +113,9 @@ class Info:
     def update_menu(self):
         """Update info menu"""
         for id in self.ids:
-            if id in self.active and self.active[id]:
-                self.window.ui.menu['info.' + id].setChecked(True)
-            else:
-                self.window.ui.menu['info.' + id].setChecked(False)
+            item = 'info.' + id
+            if item in self.window.ui.menu:
+                if id in self.active and self.active[id]:
+                    self.window.ui.menu[item].setChecked(True)
+                else:
+                    self.window.ui.menu[item].setChecked(False)

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/core/audio/capture.py

@@ -333,10 +333,10 @@ class Capture:
         :return: True if working
         """
         import pyaudio
+        p = pyaudio.PyAudio()
         try:
             rate = 44100
             channels = 1
-            p = pyaudio.PyAudio()
             stream = p.open(format=pyaudio.paInt16,
                             channels=channels,
                             rate=rate,
@@ -346,4 +346,5 @@ class Capture:
             p.terminate()
             return True
         except Exception as e:
+            p.terminate()
             return False

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/data/config/config.json

@@ -1,8 +1,8 @@
 {
   "__meta__": {
-    "version": "2.4.52",
-    "app.version": "2.4.52",
-    "updated_at": "2025-01-17T00:00:00"
+    "version": "2.4.54",
+    "app.version": "2.4.54",
+    "updated_at": "2025-01-18T00:00:00"
   },
   "access.audio.event.speech": false,
   "access.audio.event.speech.disabled": [],

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/data/config/models.json

@@ -1,8 +1,8 @@
 {
     "__meta__": {
-        "version": "2.4.52",
-        "app.version": "2.4.52",
-        "updated_at": "2025-01-17T00:00:00"
+        "version": "2.4.54",
+        "app.version": "2.4.54",
+        "updated_at": "2025-01-18T00:00:00"
     },
     "items": {
         "claude-3-5-sonnet-20240620": {

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/data/config/modes.json

@@ -1,8 +1,8 @@
 {
     "__meta__": {
-        "version": "2.4.52",
-        "app.version": "2.4.52",
-        "updated_at": "2025-01-17T00:00:00"
+        "version": "2.4.54",
+        "app.version": "2.4.54",
+        "updated_at": "2025-01-18T00:00:00"
     },
     "items": {
         "chat": {

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/plugin/audio_input/simple.py

@@ -61,6 +61,19 @@ class Simple:
 
         :param force: True to force recording
         """
+        # display snap warning if not displayed yet
+        if (not self.plugin.window.core.config.get("audio.input.snap", False)
+                or not self.plugin.window.core.config.has("audio.input.snap")):
+            if self.plugin.window.core.platforms.is_snap():
+                self.plugin.window.ui.dialogs.open(
+                    'snap_audio_input',
+                    width=400,
+                    height=200
+                )
+                self.plugin.window.core.config.set("audio.input.snap", True)
+                self.plugin.window.core.config.save()
+                return
+
         # enable continuous mode if notepad tab is active
         self.plugin.window.core.audio.capture.stop_callback = self.on_stop
         continuous_enabled = self.plugin.window.core.config.get('audio.input.continuous', False)
@@ -71,8 +84,7 @@ class Simple:
 
         try:
             # stop audio output if playing
-            if self.plugin.window.controller.audio.is_playing():
-                self.plugin.window.controller.audio.stop_output()
+            self.plugin.window.controller.audio.stop_output()
 
             # set audio volume bar
             self.plugin.window.core.audio.capture.set_bar(

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/plugin/audio_input/worker.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.11.26 19:00:00                  #
+# Updated Date: 2025.01.18 03:00:00                  #
 # ================================================== #
 
 import os.path
@@ -16,6 +16,7 @@ import audioop
 
 from PySide6.QtCore import Slot, Signal
 
+from pygpt_net.core.tabs import Tab
 from pygpt_net.utils import trans
 from pygpt_net.plugin.base.worker import BaseWorker, BaseSignals
 
@@ -76,10 +77,12 @@ class Worker(BaseWorker):
                 self.status(trans('audio.speak.wait'))
 
                 # if multimodal audio, then only return path to audio file and do not transcribe
-                if self.plugin.window.controller.chat.audio.enabled():
-                    self.signals.on_realtime.emit(self.path)
-                    self.status('')
-                    return
+                tab = self.window.controller.ui.tabs.get_current_tab()
+                if tab.type == Tab.TAB_CHAT:
+                    if self.plugin.window.controller.chat.audio.enabled():
+                        self.signals.on_realtime.emit(self.path)
+                        self.status('')
+                        return
 
                 # transcribe audio
                 transcript = self.plugin.get_provider().transcribe(self.path)

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/plugin/audio_output/__init__.py

@@ -6,9 +6,11 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 18:00:00                  #
+# Updated Date: 2025.01.18 03:00:00                  #
 # ================================================== #
 
+from typing import Any
+
 from PySide6.QtCore import Slot
 
 from pygpt_net.core.types import MODE_AUDIO
@@ -35,6 +37,7 @@ class Plugin(BasePlugin):
         self.use_locale = True
         self.output_file = "output.mp3"
         self.config = Config(self)
+        self.worker = None
 
     def init_options(self):
         """Initialize options"""
@@ -171,10 +174,12 @@ class Plugin(BasePlugin):
         # check for audio read allowed. Prevents reading audio in commands, results, etc.
         if name == Event.CTX_AFTER:
             if not ctx.audio_read_allowed():
-                return # abort if audio read is not allowed (commands, results, etc.)
-        
+                return  # abort if audio read is not allowed (commands, results, etc.)
+
         try:
             if text is not None and len(text) > 0:
+                self.stop_audio()
+
                 worker = Worker()
                 worker.from_defaults(self)
                 worker.ctx = ctx
@@ -185,13 +190,16 @@ class Plugin(BasePlugin):
 
                 # signals
                 worker.signals.playback.connect(self.handle_playback)
+                worker.signals.error_playback.connect(self.handle_playback_error)
                 worker.signals.stop.connect(self.handle_stop)
+                worker.signals.volume_changed.connect(self.handle_volume)
 
                 worker.run_async()
+                self.worker = worker
 
                 # only for manual reading
                 if name == Event.AUDIO_READ_TEXT:
-                    self.window.controller.audio.on_begin(worker.text)
+                    self.window.controller.audio.on_begin(self.worker.text)
 
         except Exception as e:
             self.error(e)
@@ -202,8 +210,10 @@ class Plugin(BasePlugin):
 
         :param ctx: CtxItem
         :param event: Event
-        """     
+        """
         try:
+            self.stop_audio()
+
             worker = Worker()
             worker.from_defaults(self)
             worker.audio_file = event.data["audio_file"]
@@ -211,9 +221,12 @@ class Plugin(BasePlugin):
 
             # signals
             worker.signals.playback.connect(self.handle_playback)
+            worker.signals.error_playback.connect(self.handle_playback_error)
             worker.signals.stop.connect(self.handle_stop)
+            worker.signals.volume_changed.connect(self.handle_volume)
 
             worker.run_async()
+            self.worker = worker
 
         except Exception as e:
             self.error(e)
@@ -250,22 +263,44 @@ class Plugin(BasePlugin):
 
         Stop playing the audio
         """
-        if self.playback is not None:
-            self.playback.stop()
-            self.playback = None
+        if self.worker is not None:
+            self.worker.stop()
+        self.handle_volume(0.0)
 
-    @Slot(object, str)
-    def handle_playback(self, playback, event: str):
+    @Slot(object)
+    def handle_playback_error(self, err: Any):
+        """
+        Send error message to logger and alert dialog
+
+        :param err: error message
+        """
+        self.error(err)
+        if self.window.core.platforms.is_snap():
+            self.window.ui.dialogs.open(
+                'snap_audio_output',
+                width=400,
+                height=200
+            )
+
+    @Slot(str)
+    def handle_playback(self, event: str):
         """
         Handle thread playback object
 
-        :param playback: playback object
         :param event: event name
         """
-        self.playback = playback
         self.window.controller.audio.on_play(event)
 
     @Slot()
     def handle_stop(self):
         """Handle thread playback stop"""
         self.stop_audio()
+
+    @Slot(float)
+    def handle_volume(self, volume: float):
+        """
+        Handle thread playback volume
+
+        :param volume: volume level
+        """
+        self.window.ui.plugin_addon['audio.output.bar'].setLevel(volume)

pygpt_net-2.4.54/src/pygpt_net/plugin/audio_output/worker.py

@@ -0,0 +1,199 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+# ================================================== #
+# This file is a part of PYGPT package               #
+# Website: https://pygpt.net                         #
+# GitHub:  https://github.com/szczyglis-dev/py-gpt   #
+# MIT License                                        #
+# Created By  : Marcin Szczygliński                  #
+# Updated Date: 2025.01.18 03:00:00                  #
+# ================================================== #
+
+import time
+import pyaudio
+import wave
+import numpy as np
+import io
+from pydub import AudioSegment
+
+from PySide6.QtCore import Slot, Signal
+
+from pygpt_net.plugin.base.worker import BaseWorker, BaseSignals
+
+
+class WorkerSignals(BaseSignals):
+    volume_changed = Signal(float)  # Emits the volume level
+    playback = Signal(str)
+    stop = Signal()
+    error_playback = Signal(object)
+
+
+class Worker(BaseWorker):
+    def __init__(self, *args, **kwargs):
+        super(Worker, self).__init__()
+        self.signals = WorkerSignals()
+        self.args = args
+        self.kwargs = kwargs
+        self.plugin = None
+        self.text = None
+        self.event = None
+        self.cache_file = None  # path to cache file
+        self.mode = "generate"  # generate|playback
+        self.audio_file = None
+        self.is_stopped = False
+        self.p = None
+        self.stream = None
+
+    @Slot()
+    def run(self):
+        try:
+            if self.mode == "generate":
+                self.generate()
+            elif self.mode == "playback":
+                self.play()
+        except Exception as e:
+            self.error(e)
+
+    def generate(self):
+        """Generate and play audio file"""
+        if self.text is None or self.text == "":
+            time.sleep(0.2)  # wait
+            return
+        path = self.plugin.get_provider().speech(self.text)
+        if path:
+            self.play_audio(path)
+            # store in cache if enabled
+            if self.cache_file:
+                self.cache_audio_file(path, self.cache_file)
+
+    def play(self):
+        """Play audio file only"""
+        if self.audio_file:
+            self.play_audio(self.audio_file)
+
+    def play_audio(self, audio_file: str):
+        """
+        Play audio file using PyAudio
+
+        :param audio_file: audio file path
+        """
+        try:
+            self.signals.playback.emit(self.event)
+            audio = AudioSegment.from_file(audio_file)
+            audio = audio.set_frame_rate(44100)  # resample to 44.1 kHz
+            wav_io = io.BytesIO()
+            audio.export(wav_io, format='wav')
+            wav_io.seek(0)
+            wf = wave.open(wav_io, 'rb')
+            self.p = pyaudio.PyAudio()
+            self.stream = self.p.open(format=self.p.get_format_from_width(wf.getsampwidth()),
+                                      channels=wf.getnchannels(),
+                                      rate=wf.getframerate(),
+                                      output=True)
+
+            sample_width = wf.getsampwidth()
+            format = self.p.get_format_from_width(sample_width)
+
+            if format == pyaudio.paInt8:
+                dtype = np.int8
+                max_value = 2 ** 7 - 1  # 127
+                offset = 0
+            elif format == pyaudio.paInt16:
+                dtype = np.int16
+                max_value = 2 ** 15 - 1  # 32767
+                offset = 0
+            elif format == pyaudio.paInt32:
+                dtype = np.int32
+                max_value = 2 ** 31 - 1  # 2147483647
+                offset = 0
+            elif format == pyaudio.paUInt8:
+                dtype = np.uint8
+                max_value = 2 ** 8 - 1  # 255
+                offset = 128  # center unsigned data
+            else:
+                raise ValueError(f"Unsupported format: {format}")
+
+            chunk_size = 512
+            data = wf.readframes(chunk_size)
+
+            while data != b'' and not self.is_stopped:
+                self.stream.write(data)
+
+                audio_data = np.frombuffer(data, dtype=dtype)
+                if len(audio_data) > 0:
+                    audio_data = audio_data.astype(np.float32)
+                    if dtype == np.uint8:
+                        audio_data -= offset
+
+                    # compute RMS
+                    rms = np.sqrt(np.mean(audio_data ** 2))
+
+                    if rms > 0:
+                        # RMS to decibels
+                        db = 20 * np.log10(rms / max_value)
+
+                        # define minimum and maximum dB levels
+                        min_db = -60  # adjust as needed
+                        max_db = 0
+
+                        # clamp the db value to the range [min_db, max_db]
+                        if db < min_db:
+                            db = min_db
+                        elif db > max_db:
+                            db = max_db
+
+                        # map decibel value to volume percentage
+                        volume_percentage = ((db - min_db) / (max_db - min_db)) * 100
+                    else:
+                        volume_percentage = 0
+
+                    # emit volume signal
+                    self.signals.volume_changed.emit(volume_percentage)
+                else:
+                    # if empty audio_data
+                    self.signals.volume_changed.emit(0)
+
+                data = wf.readframes(chunk_size)
+
+            # close the stream
+            if self.stream is not None:
+                if self.stream.is_active():
+                    self.stream.stop_stream()
+                self.stream.close()
+            if self.p is not None:
+                self.p.terminate()
+            wf.close()
+            self.signals.volume_changed.emit(0)
+        except Exception as e:
+            self.signals.volume_changed.emit(0)
+            self.signals.error_playback.emit(e)
+
+    def cache_audio_file(self, src: str, dst: str):
+        """
+        Store audio file in cache
+
+        :param src: source path
+        :param dst: destination path
+        """
+        import shutil
+        try:
+            shutil.copy(src, dst)
+            # print("Cached audio file:", dst)
+        except Exception as e:
+            self.error(e)
+
+    def send(self, playback):
+        """
+        Send playback object to main thread
+
+        :param playback: playback object
+        """
+        self.signals.playback.emit(playback, self.event)
+
+    def stop_playback(self):
+        """Stop audio playback"""
+        self.stop()
+
+    def stop(self):
+        """Send stop signal to main thread"""
+        self.is_stopped = True

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/ui/__init__.py

@@ -91,7 +91,8 @@ class UI:
 
         # FIRST RUN: initial sizes if not set yet
         if not self.window.core.config.has("layout.splitters") \
-            or self.window.core.config.get("layout.splitters") == {}:
+            or self.window.core.config.get("layout.splitters") == {}\
+            or self.window.core.config.get("license.accepted") == False:
             self.set_initial_size()
 
         # menus

{pygpt_net-2.4.52 → pygpt_net-2.4.54}/src/pygpt_net/ui/dialog/snap.py

@@ -6,10 +6,10 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.02.26 22:00:00                  #
+# Updated Date: 2025.01.18 05:00:00                  #
 # ================================================== #
 
-from pygpt_net.ui.widget.dialog.snap import SnapDialogCamera, SnapDialogAudioInput
+from pygpt_net.ui.widget.dialog.snap import SnapDialogCamera, SnapDialogAudioInput, SnapDialogAudioOutput
 
 
 class Snap:
@@ -25,3 +25,4 @@ class Snap:
         """Setup snap dialog"""
         self.window.ui.dialog['snap_camera'] = SnapDialogCamera(self.window)
         self.window.ui.dialog['snap_audio_input'] = SnapDialogAudioInput(self.window)
+        self.window.ui.dialog['snap_audio_output'] = SnapDialogAudioOutput(self.window)