diff --git a/config.py b/config.py index ce6bfd1f..9e17df87 100644 --- a/config.py +++ b/config.py @@ -6,7 +6,7 @@ from json import dump as json_dump import tkinter as tk from tkinter import font from models.translation.translation_languages import translation_lang -from models.transcription.transcription_utils import getInputDevices, getDefaultInputDevice +from models.transcription.transcription_utils import getInputDevices, getDefaultInputDevice, getOutputDevices, getDefaultOutputDevice from models.transcription.transcription_languages import transcription_lang from utils import generatePercentageStringsList, isUniqueStrings @@ -537,6 +537,17 @@ class Config: self._INPUT_MIC_WORD_FILTER = sorted(set(value), key=value.index) saveJson(self.PATH_CONFIG, inspect.currentframe().f_code.co_name, value) + @property + @json_serializable('CHOICE_SPEAKER_DEVICE') + def CHOICE_SPEAKER_DEVICE(self): + return self._CHOICE_SPEAKER_DEVICE + + @CHOICE_SPEAKER_DEVICE.setter + def CHOICE_SPEAKER_DEVICE(self, value): + if value in [device["name"] for device in getOutputDevices()]: + self._CHOICE_SPEAKER_DEVICE = value + saveJson(self.PATH_CONFIG, inspect.currentframe().f_code.co_name, value) + @property @json_serializable('INPUT_SPEAKER_ENERGY_THRESHOLD') def INPUT_SPEAKER_ENERGY_THRESHOLD(self): @@ -957,6 +968,7 @@ class Config: self._INPUT_MIC_PHRASE_TIMEOUT = 3 self._INPUT_MIC_MAX_PHRASES = 10 self._INPUT_MIC_WORD_FILTER = [] + self._CHOICE_SPEAKER_DEVICE = getDefaultOutputDevice()["device"]["name"] self._INPUT_SPEAKER_ENERGY_THRESHOLD = 300 self._INPUT_SPEAKER_DYNAMIC_ENERGY_THRESHOLD = False self._INPUT_SPEAKER_RECORD_TIMEOUT = 3 diff --git a/model.py b/model.py index 2af4135a..265c491d 100644 --- a/model.py +++ b/model.py @@ -16,7 +16,7 @@ import webbrowser from typing import Callable from flashtext import KeywordProcessor from models.translation.translation_translator import Translator -from models.transcription.transcription_utils import getInputDevices, getDefaultOutputDevice +from models.transcription.transcription_utils import getInputDevices, getOutputDevices from models.osc.osc_tools import sendTyping, sendMessage, sendTestAction, receiveOscParameters from models.transcription.transcription_recorder import SelectedMicEnergyAndAudioRecorder, SelectedSpeakerEnergyAndAudioRecorder from models.transcription.transcription_recorder import SelectedMicEnergyRecorder, SelectedSpeakerEnergyRecorder @@ -305,12 +305,8 @@ class Model: return [device["name"] for device in getInputDevices()[config.CHOICE_MIC_HOST]] @staticmethod - def getInputDefaultDevice(): - return [device["name"] for device in getInputDevices()[config.CHOICE_MIC_HOST]][0] - - @staticmethod - def getOutputDefaultDevice(): - return getDefaultOutputDevice()["name"] + def getListOutputDevice(): + return [device["name"] for device in getOutputDevices()] def startMicTranscript(self, fnc, error_fnc=None): if config.CHOICE_MIC_HOST == "NoHost" or config.CHOICE_MIC_DEVICE == "NoDevice": @@ -322,14 +318,14 @@ class Model: mic_audio_queue = Queue() # mic_energy_queue = Queue() - device = [device for device in getInputDevices()[config.CHOICE_MIC_HOST] if device["name"] == config.CHOICE_MIC_DEVICE][0] + mic_device = [device for device in getInputDevices()[config.CHOICE_MIC_HOST] if device["name"] == config.CHOICE_MIC_DEVICE][0] record_timeout = config.INPUT_MIC_RECORD_TIMEOUT phase_timeout = config.INPUT_MIC_PHRASE_TIMEOUT if record_timeout > phase_timeout: record_timeout = phase_timeout self.mic_audio_recorder = SelectedMicEnergyAndAudioRecorder( - device=device, + device=mic_device, energy_threshold=config.INPUT_MIC_ENERGY_THRESHOLD, dynamic_energy_threshold=config.INPUT_MIC_DYNAMIC_ENERGY_THRESHOLD, record_timeout=record_timeout, @@ -422,8 +418,7 @@ class Model: self.mic_energy_recorder = None def startSpeakerTranscript(self, fnc, error_fnc=None): - speaker_device = getDefaultOutputDevice() - if speaker_device["name"] == "NoDevice": + if config.CHOICE_SPEAKER_DEVICE == "NoDevice": try: error_fnc() except Exception: @@ -432,6 +427,8 @@ class Model: speaker_audio_queue = Queue() # speaker_energy_queue = Queue() + print([device for device in getOutputDevices() if device["name"] == config.CHOICE_SPEAKER_DEVICE]) + speaker_device = [device for device in getOutputDevices() if device["name"] == config.CHOICE_SPEAKER_DEVICE][0] record_timeout = config.INPUT_SPEAKER_RECORD_TIMEOUT phase_timeout = config.INPUT_SPEAKER_PHRASE_TIMEOUT if record_timeout > phase_timeout: @@ -498,8 +495,7 @@ class Model: # self.speaker_get_energy = None def startCheckSpeakerEnergy(self, fnc, end_fnc, error_fnc=None): - speaker_device = getDefaultOutputDevice() - if speaker_device["name"] == "NoDevice": + if config.CHOICE_SPEAKER_DEVICE == "NoDevice": try: error_fnc() except Exception: @@ -516,6 +512,7 @@ class Model: sleep(0.01) speaker_energy_queue = Queue() + speaker_device = [device for device in getOutputDevices() if device["name"] == config.CHOICE_SPEAKER_DEVICE][0] self.speaker_energy_recorder = SelectedSpeakerEnergyRecorder(speaker_device) self.speaker_energy_recorder.recordIntoQueue(speaker_energy_queue) self.speaker_energy_plot_progressbar = threadFnc(sendSpeakerEnergy, end_fnc=end_fnc) diff --git a/models/transcription/transcription_utils.py b/models/transcription/transcription_utils.py index f40defeb..e4a8ecc8 100644 --- a/models/transcription/transcription_utils.py +++ b/models/transcription/transcription_utils.py @@ -23,12 +23,32 @@ def getDefaultInputDevice(): for host_index in range(0, p.get_host_api_count()): host = p.get_host_api_info_by_index(host_index) - for device_index in range(0, p. get_host_api_info_by_index(host_index)['deviceCount']): + for device_index in range(0, p.get_host_api_info_by_index(host_index)['deviceCount']): device = p.get_device_info_by_host_api_device_index(host_index, device_index) if device["index"] == defaultInputDevice: return {"host": host, "device": device} return {"host": {"name": "NoHost"}, "device": {"name": "NoDevice"}} +def getOutputDevices(): + devices = [] + with PyAudio() as p: + wasapi_info = p.get_host_api_info_by_type(paWASAPI) + defaultOutputDevice = wasapi_info["defaultOutputDevice"] + + for host_index in range(0, p.get_host_api_count()): + for device_index in range(0, p.get_host_api_info_by_index(host_index)['deviceCount']): + device = p.get_device_info_by_host_api_device_index(host_index, device_index) + if device["index"] == defaultOutputDevice: + default_speakers = device + if not default_speakers["isLoopbackDevice"]: + for loopback in p.get_loopback_device_info_generator(): + if default_speakers["name"] in loopback["name"]: + devices.append(loopback) + + if len(devices) == 0: + devices = [{"name": "NoDevice"}] + return devices + def getDefaultOutputDevice(): with PyAudio() as p: wasapi_info = p.get_host_api_info_by_type(paWASAPI) @@ -42,6 +62,9 @@ def getDefaultOutputDevice(): if not default_speakers["isLoopbackDevice"]: for loopback in p.get_loopback_device_info_generator(): if default_speakers["name"] in loopback["name"]: - default_device = loopback - return default_device - return {"name":"NoDevice"} \ No newline at end of file + return {"device": loopback} + return {"device": {"name": "NoDevice"}} + +if __name__ == "__main__": + print("getOutputDevices():", getOutputDevices()) + print("getDefaultOutputDevice()", getDefaultOutputDevice()) \ No newline at end of file