diff --git a/extensions/silero_tts/requirements.txt b/extensions/silero_tts/requirements.txt index b4444306..f2f0bff5 100644 --- a/extensions/silero_tts/requirements.txt +++ b/extensions/silero_tts/requirements.txt @@ -4,4 +4,3 @@ pydub PyYAML torch torchaudio -simpleaudio diff --git a/extensions/silero_tts/script.py b/extensions/silero_tts/script.py index 334b02b9..b66963e2 100644 --- a/extensions/silero_tts/script.py +++ b/extensions/silero_tts/script.py @@ -4,7 +4,6 @@ import gradio as gr import torch import modules.shared as shared -import simpleaudio as sa torch._C._jit_set_profiling_mode(False) @@ -15,13 +14,16 @@ params = { 'model_id': 'v3_en', 'sample_rate': 48000, 'device': 'cpu', - 'max_wavs': -1, - 'autoplay': True, 'show_text': True, + 'autoplay': True, + 'voice_pitch': 'medium', + 'voice_speed': 'medium', } current_params = params.copy() voices_by_gender = ['en_99', 'en_45', 'en_18', 'en_117', 'en_49', 'en_51', 'en_68', 'en_0', 'en_26', 'en_56', 'en_74', 'en_5', 'en_38', 'en_53', 'en_21', 'en_37', 'en_107', 'en_10', 'en_82', 'en_16', 'en_41', 'en_12', 'en_67', 'en_61', 'en_14', 'en_11', 'en_39', 'en_52', 'en_24', 'en_97', 'en_28', 'en_72', 'en_94', 'en_36', 'en_4', 'en_43', 'en_88', 'en_25', 'en_65', 'en_6', 'en_44', 'en_75', 'en_91', 'en_60', 'en_109', 'en_85', 'en_101', 'en_108', 'en_50', 'en_96', 'en_64', 'en_92', 'en_76', 'en_33', 'en_116', 'en_48', 'en_98', 'en_86', 'en_62', 'en_54', 'en_95', 'en_55', 'en_111', 'en_3', 'en_83', 'en_8', 'en_47', 'en_59', 'en_1', 'en_2', 'en_7', 'en_9', 'en_13', 'en_15', 'en_17', 'en_19', 'en_20', 'en_22', 'en_23', 'en_27', 'en_29', 'en_30', 'en_31', 'en_32', 'en_34', 'en_35', 'en_40', 'en_42', 'en_46', 'en_57', 'en_58', 'en_63', 'en_66', 'en_69', 'en_70', 'en_71', 'en_73', 'en_77', 'en_78', 'en_79', 'en_80', 'en_81', 'en_84', 'en_87', 'en_89', 'en_90', 'en_93', 'en_100', 'en_102', 'en_103', 'en_104', 'en_105', 'en_106', 'en_110', 'en_112', 'en_113', 'en_114', 'en_115'] -wav_idx = 0 +voice_pitches = ['x-low', 'low', 'medium', 'high', 'x-high'] +voice_speeds = ['x-slow', 'slow', 'medium', 'fast', 'x-fast'] +last_msg_id = 0 #Used for making text xml compatible, needed for voice pitch and speed control table = str.maketrans({ @@ -55,6 +57,14 @@ def input_modifier(string): This function is applied to your text inputs before they are fed into the model. """ + #remove autoplay from previous + if len(shared.history['internal'])>0: + [text, reply] = shared.history['internal'][-1] + [visible_text, visible_reply] = shared.history['visible'][-1] + rep_clean = reply.replace('controls autoplay>','controls>') + vis_rep_clean = visible_reply.replace('controls autoplay>','controls>') + shared.history['internal'][-1] = [text, rep_clean] + shared.history['visible'][-1] = [visible_text, vis_rep_clean] return string @@ -63,7 +73,7 @@ def output_modifier(string): This function is applied to the model outputs. """ - global wav_idx, model, current_params + global model, current_params for i in params: if params[i] != current_params[i]: @@ -81,44 +91,31 @@ def output_modifier(string): string = string.replace('\n', ' ') string = string.strip() - auto_playable=True + silent_string = False #Used to prevent unnecessary audio file generation if string == '': string = 'empty reply, try regenerating' - auto_playable=False - + silent_string = True #x-slow, slow, medium, fast, x-fast #x-low, low, medium, high, x-high - prosody='' + pitch = params['voice_pitch'] + speed = params['voice_speed'] + prosody=f'' string =''+prosody+xmlesc(string)+'' - - output_file = Path(f'extensions/silero_tts/outputs/{wav_idx:06d}.wav') - autoplay_str = '' - if not shared.still_streaming: + + current_msg_id=len(shared.history['visible'])#check length here, since output_modifier can run many times on the same message + output_file = Path(f'extensions/silero_tts/outputs/{shared.character}_{current_msg_id:06d}.wav') + if not shared.still_streaming and not silent_string: model.save_wav(ssml_text=string, speaker=params['speaker'], sample_rate=int(params['sample_rate']), audio_path=str(output_file)) - #diabled until autoplay doesn't run on previous messages - #autoplay = 'autoplay' if (params['autoplay'] and auto_playable) else '' - string = f'\n\n' + string = f'\n\n' else: - #placeholder so text doesnt shift around so much - string =f'\n\n' - - #reset if too many wavs. set max to -1 for unlimited. - if wav_idx < params['max_wavs'] or params['max_wavs'] < 0: - #only increment if starting a new stream, else replace during streaming. - if not shared.still_streaming: - wav_idx += 1 - else: - wav_idx = 0 + #placeholder so text doesn't shift around so much + string ='\n\n' if params['show_text']: + #string+=f'*[{current_msg_id}]:*'+orig_string #Debug, looks like there is a delay in "current_msg_id" being updated when switching characters (updates after new message sent). Can't find the source. "shared.character" is updating properly. string+=orig_string - if params['autoplay'] == True and auto_playable and not shared.still_streaming: - stop_autoplay() - wave_obj = sa.WaveObject.from_wave_file(output_file.as_posix()) - wave_obj.play() - return string def bot_prefix_modifier(string): @@ -130,20 +127,20 @@ def bot_prefix_modifier(string): return string -def stop_autoplay(): - sa.stop_all() - def ui(): # Gradio elements - activate = gr.Checkbox(value=params['activate'], label='Activate TTS') - show_text = gr.Checkbox(value=params['show_text'], label='Show message text under audio player') - autoplay = gr.Checkbox(value=params['autoplay'], label='Play TTS automatically') - stop_audio = gr.Button("Stop Auto-Play") - voice = gr.Dropdown(value=params['speaker'], choices=voices_by_gender, label='TTS voice') + with gr.Accordion("Silero TTS"): + activate = gr.Checkbox(value=params['activate'], label='Activate TTS') + show_text = gr.Checkbox(value=params['show_text'], label='Show message text under audio player') + autoplay = gr.Checkbox(value=params['autoplay'], label='Play TTS automatically') + voice = gr.Dropdown(value=params['speaker'], choices=voices_by_gender, label='TTS voice') + v_pitch = gr.Dropdown(value=params['voice_pitch'], choices=voice_pitches, label='Voice pitch') + v_speed = gr.Dropdown(value=params['voice_speed'], choices=voice_speeds, label='Voice speed') # Event functions to update the parameters in the backend activate.change(lambda x: params.update({"activate": x}), activate, None) - autoplay.change(lambda x: params.update({"autoplay": x}), autoplay, None) show_text.change(lambda x: params.update({"show_text": x}), show_text, None) - stop_audio.click(stop_autoplay) + autoplay.change(lambda x: params.update({"autoplay": x}), autoplay, None) voice.change(lambda x: params.update({"speaker": x}), voice, None) + v_pitch.change(lambda x: params.update({"voice_pitch": x}), v_pitch, None) + v_speed.change(lambda x: params.update({"voice_speed": x}), v_speed, None)