pygpt-net 2.4.53__py3-none-any.whl → 2.4.55__py3-none-any.whl

CHANGELOG.md

@@ -1,5 +1,15 @@
 # CHANGELOG
 
+## 2.4.55 (2025-01-18)
+
+- Added a new option in settings: Audio -> Recording timeout.
+- Added a new option in settings: Audio -> Enable timeout in continuous mode.
+
+## 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

README.md

@@ -2,7 +2,7 @@
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.4.53** | build: **2025.01.17** | Python: **>=3.10, <3.13**
+Release: **2.4.55** | 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 +3960,16 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.4.55 (2025-01-18)**
+
+- Added a new option in settings: Audio -> Recording timeout.
+- Added a new option in settings: Audio -> Enable timeout in continuous mode.
+
+**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

pygpt_net/CHANGELOG.txt

@@ -1,3 +1,13 @@
+2.4.55 (2025-01-18)
+
+- Added a new option in settings: Audio -> Recording timeout.
+- Added a new option in settings: Audio -> Enable timeout in continuous mode.
+
+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

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 16:00:00                  #
 # ================================================== #
 
 __author__ = "Marcin Szczygliński"
 __copyright__ = "Copyright 2025, Marcin Szczygliński"
 __credits__ = ["Marcin Szczygliński"]
 __license__ = "MIT"
-__version__ = "2.4.53"
-__build__ = "2025.01.17"
+__version__ = "2.4.55"
+__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/controller/lang/mapping.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: 2025.01.18 16:00:00                  #
 # ================================================== #
 
 from typing import Dict
@@ -145,6 +145,7 @@ class Mapping:
         nodes["agent.iterations.label"] = "toolbox.agent.iterations.label"
         nodes["agent.auto_stop"] = "toolbox.agent.auto_stop.label"
         nodes["agent.continue"] = "toolbox.agent.continue.label"
+        nodes['layout.split'] = "layout.split"
         # nodes["indexes.new"] = "idx.new"
 
         # input

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/data/config/config.json

@@ -1,8 +1,8 @@
 {
   "__meta__": {
-    "version": "2.4.53",
-    "app.version": "2.4.53",
-    "updated_at": "2025-01-17T00:00:00"
+    "version": "2.4.55",
+    "app.version": "2.4.55",
+    "updated_at": "2025-01-18T00:00:00"
   },
   "access.audio.event.speech": false,
   "access.audio.event.speech.disabled": [],
@@ -78,6 +78,8 @@
   "audio.input.device": "0",
   "audio.input.rate": 44100,
   "audio.input.stop_interval": 10,
+  "audio.input.timeout": 120,
+  "audio.input.timeout.continuous": false,
   "audio.transcribe.convert_video": true,
   "context_threshold": 200,
   "cmd": false,

pygpt_net/data/config/models.json

@@ -1,8 +1,8 @@
 {
     "__meta__": {
-        "version": "2.4.53",
-        "app.version": "2.4.53",
-        "updated_at": "2025-01-17T00:00:00"
+        "version": "2.4.55",
+        "app.version": "2.4.55",
+        "updated_at": "2025-01-18T00:00:00"
     },
     "items": {
         "claude-3-5-sonnet-20240620": {

pygpt_net/data/config/modes.json

@@ -1,8 +1,8 @@
 {
     "__meta__": {
-        "version": "2.4.53",
-        "app.version": "2.4.53",
-        "updated_at": "2025-01-17T00:00:00"
+        "version": "2.4.55",
+        "app.version": "2.4.55",
+        "updated_at": "2025-01-18T00:00:00"
     },
     "items": {
         "chat": {

pygpt_net/data/config/settings.json

@@ -1136,6 +1136,30 @@
         "step": 1,
         "advanced": false
     },
+    "audio.input.timeout": {
+        "section": "audio",
+        "type": "int",
+        "slider": false,
+        "label": "settings.audio.input.timeout",
+        "description": "settings.audio.input.timeout.desc",
+        "value": 120,
+        "min": 0,
+        "multiplier": 1,
+        "step": 1,
+        "advanced": false
+    },
+    "audio.input.timeout.continuous": {
+        "section": "audio",
+        "type": "bool",
+        "slider": false,
+        "label": "settings.audio.input.timeout.continuous",
+        "value": false,
+        "min": null,
+        "max": null,
+        "multiplier": null,
+        "step": null,
+        "advanced": false
+    },
     "llama.idx.list": {
         "section": "llama-index",
         "type": "dict",

pygpt_net/data/locale/locale.de.ini

@@ -777,6 +777,9 @@ settings.audio.input.rate = Abtastrate
 settings.audio.input.rate.desc = Abtastrate, Standard: 44100
 settings.audio.input.stop_interval = Intervall für kontinuierliche automatische Transkription
 settings.audio.input.stop_interval.desc = Intervall in Sekunden für automatisches Transkribieren eines Audioabschnitts, Standard: 10
+settings.audio.input.timeout = Aufnahme-Zeitüberschreitung
+settings.audio.input.timeout.desc = Zeitüberschreitung (Sekunden) für automatische Stop-Aufzeichnung, 0 zum Deaktivieren, Standard: 120
+settings.audio.input.timeout.continuous = Zeitüberschreitung im Dauerbetrieb aktivieren
 settings.check_updates = Beim Start nach Updates suchen
 settings.check_updates.bg = Im Hintergrund nach Updates suchen
 settings.cmd.field.desc = Aktivieren Sie das `{cmd}`-Werkzeug.

pygpt_net/data/locale/locale.en.ini

@@ -948,6 +948,9 @@ settings.audio.input.rate = Sampling Rate
 settings.audio.input.rate.desc = Sampling rate, default: 44100
 settings.audio.input.stop_interval = Continuous recording auto-transcribe interval
 settings.audio.input.stop_interval.desc = Interval in seconds for auto-transcribe audio chunk, default: 10
+settings.audio.input.timeout = Recording timeout
+settings.audio.input.timeout.desc = Timeout (seconds) for auto-stop recording, 0 to disable, default: 120
+settings.audio.input.timeout.continuous = Enable timeout in continuous mode
 settings.check_updates = Check for Updates on start
 settings.check_updates.bg = Check for Updates in the background
 settings.cmd.config.collapse = JSON params

pygpt_net/data/locale/locale.es.ini

@@ -776,6 +776,9 @@ settings.audio.input.rate = Taux d'échantillonnage
 settings.audio.input.rate.desc = Taux d'échantillonnage, par défaut : 44100
 settings.audio.input.stop_interval = Intervalo de transcripción automática continua
 settings.audio.input.stop_interval.desc = Intervalo en segundos para transcribir automáticamente el fragmento de audio, predeterminado: 10
+settings.audio.input.timeout = Tiempo de espera de grabación
+settings.audio.input.timeout.desc = Tiempo de espera (segundos) para detener la grabación automáticamente, 0 para deshabilitar, por defecto: 120
+settings.audio.input.timeout.continuous = Habilitar tiempo de espera en modo continuo
 settings.check_updates = Buscar actualizaciones al iniciar
 settings.check_updates.bg = Buscar actualizaciones en segundo plano
 settings.cmd.field.desc = Habilitar la herramienta `{cmd}`.

pygpt_net/data/locale/locale.fr.ini

@@ -776,6 +776,9 @@ settings.audio.input.rate = Taux d'échantillonnage
 settings.audio.input.rate.desc = Taux d'échantillonnage, par défaut : 44100
 settings.audio.input.stop_interval = Intervalle d'auto-transcription enregistrement continu
 settings.audio.input.stop_interval.desc = Intervalle en secondes pour la transcription automatique d'un morceau audio, par défaut : 10
+settings.audio.input.timeout = Délai d'enregistrement
+settings.audio.input.timeout.desc = Délai (secondes) pour l'arrêt automatique de l'enregistrement, 0 pour désactiver, par défaut : 120
+settings.audio.input.timeout.continuous = Activer le délai en mode continu
 settings.check_updates = Vérifier les mises à jour au démarrage
 settings.check_updates.bg = Vérifier les mises à jour en arrière-plan
 settings.cmd.field.desc = Activer l'outil `{cmd}`.

pygpt_net/data/locale/locale.it.ini

@@ -777,6 +777,9 @@ settings.audio.input.rate = Frequenza di campionamento
 settings.audio.input.rate.desc = Frequenza di campionamento, predefinito: 44100
 settings.audio.input.stop_interval = Intervallo di trascrizione automatica continua
 settings.audio.input.stop_interval.desc = Intervallo in secondi per la trascrizione automatica del segmento audio, predefinito: 10
+settings.audio.input.timeout = Timeout di registrazione
+settings.audio.input.timeout.desc = Timeout (secondi) per l'arresto automatico della registrazione, 0 per disabilitare, predefinito: 120
+settings.audio.input.timeout.continuous = Abilita timeout in modalità continua
 settings.check_updates = Controlla aggiornamenti all'avvio
 settings.check_updates.bg = Controlla gli aggiornamenti in background
 settings.cmd.field.desc = Abilita lo strumento `{cmd}`.

pygpt_net/data/locale/locale.pl.ini

@@ -776,7 +776,10 @@ settings.audio.input.device.desc = Wybierz urządzenie audio dla wejścia mikrof
 settings.audio.input.rate = Częstotliwość próbkowania
 settings.audio.input.rate.desc = Częstotliwość próbkowania, domyślnie: 44100
 settings.audio.input.stop_interval = Interwał automatycznego transkrybowania nagrania ciągłego
-settings.audio.input.stop_interval.desc = Interwał w sekundach dla automatycznego transkrybowania fragmentu audio, domyślnie: 10
+settings.audio.input.stop_interval.desc = Interwał w sekundach dla automatycznego transkrybowania fragmentu audio, domyślnie: 10
+settings.audio.input.timeout = Limit czasu nagrywania
+settings.audio.input.timeout.desc = Limit czasu (w sekundach) dla automatycznego zatrzymania nagrywania, 0 aby wyłączyć, domyślnie: 120
+settings.audio.input.timeout.continuous = Włącz limit czasu w trybie ciągłym
 settings.check_updates = Sprawdź aktualizacje przy starcie
 settings.check_updates.bg = Sprawdź aktualizacje w tle
 settings.cmd.field.desc = Włącz narzędzie `{cmd}`.

pygpt_net/data/locale/locale.uk.ini

@@ -776,6 +776,9 @@ settings.audio.input.rate = Частота дискретизації
 settings.audio.input.rate.desc = Частота дискретизації, за замовчуванням: 44100
 settings.audio.input.stop_interval = Інтервал автоматичного транскрибування безперервного запису
 settings.audio.input.stop_interval.desc = Інтервал у секундах для автоматичного транскрибування фрагмента аудіо, за замовчуванням: 10
+settings.audio.input.timeout = Тайм-аут запису
+settings.audio.input.timeout.desc = Тайм-аут (в секундах) для автоматичної зупинки запису, 0 для відключення, за замовчуванням: 120
+settings.audio.input.timeout.continuous = Увімкнути тайм-аут у безперервному режимі
 settings.check_updates = Перевіряти оновлення при запуску
 settings.check_updates.bg = Перевіряти оновлення у фоновому режимі
 settings.cmd.field.desc = Увімкнути інструмент `{cmd}`.

pygpt_net/data/locale/locale.zh.ini

@@ -878,6 +878,9 @@ settings.audio.input.rate = 采样率
 settings.audio.input.rate.desc = 采样率，默认: 44100
 settings.audio.input.stop_interval = 连续录音自动转录间隔
 settings.audio.input.stop_interval.desc = 自动转录音频片段的间隔（以秒为单位），默认：10
+settings.audio.input.timeout = 录音超时
+settings.audio.input.timeout.desc = 自动停止录音的超时时间（秒），0为禁用，默认：120
+settings.audio.input.timeout.continuous = 在连续模式下启用超时
 settings.check_updates = 啟動時檢查更新
 settings.check_updates.bg = 在後台檢查更新
 settings.cmd.field.desc = 启用 `{cmd}` 工具。

pygpt_net/plugin/audio_input/simple.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.01.17 02:00:00                  #
+# Updated Date: 2025.01.18 16:00:00                  #
 # ================================================== #
 
 import os
@@ -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(
@@ -80,10 +92,14 @@ class Simple:
             )
 
             # start timeout timer to prevent infinite recording
-            if self.timer is None:
-                self.timer = QTimer()
-                self.timer.timeout.connect(self.stop_timeout)
-                self.timer.start(self.TIMEOUT_SECONDS * 1000)
+            # disable in continuous mode
+            timeout = int(self.plugin.window.core.config.get('audio.input.timeout', 120) or 0) # get timeout
+            timeout_continuous = self.plugin.window.core.config.get('audio.input.timeout.continuous', False) # enable continuous timeout
+            if timeout > 0:
+                if self.timer is None and (not continuous_enabled or timeout_continuous):
+                    self.timer = QTimer()
+                    self.timer.timeout.connect(self.stop_timeout)
+                    self.timer.start(timeout * 1000)
 
             if not force:
                 if not self.plugin.window.core.audio.capture.check_audio_input():

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/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)