From 9907bee4a45a9ce9c287b5c294f525aa8bae79b0 Mon Sep 17 00:00:00 2001 From: MetaIX <125941078+MetaIX@users.noreply.github.com> Date: Sun, 5 Mar 2023 19:04:22 -0600 Subject: [PATCH 1/3] Support for Eleven Labs TTS As per your suggestion at https://github.com/oobabooga/text-generation-webui/issues/159 here's my attempt. I'm brand new to python and github. Completely different from unreal + visual coding, so forgive my amateurish code. This essentially adds support for Eleven Labs TTS. Tested it without major issues, and I believe it's functional (hopefully). Extra requirements: elevenlabslib https://github.com/lugia19/elevenlabslib, sounddevice0.4.6, and soundfile Folder structure is the same as the SileroTTS Extension. --- extensions/silero_tts/script.py | 102 ++++++++++++++++++++------------ 1 file changed, 64 insertions(+), 38 deletions(-) diff --git a/extensions/silero_tts/script.py b/extensions/silero_tts/script.py index f697d0e2..e088ec8e 100644 --- a/extensions/silero_tts/script.py +++ b/extensions/silero_tts/script.py @@ -3,26 +3,55 @@ from pathlib import Path import gradio as gr import torch +import io +import json +import os -torch._C._jit_set_profiling_mode(False) +import requests + +from elevenlabslib.helpers import * +from elevenlabslib import * params = { 'activate': True, - 'speaker': 'en_56', - 'language': 'en', - 'model_id': 'v3_en', - 'sample_rate': 48000, - 'device': 'cpu', + 'api_key': '12345', + 'selected_voice': 'None', } -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'] +initial_voice = ['None'] wav_idx = 0 +user = ElevenLabsUser(params['api_key']) +user_info = None -def load_model(): - model, example_text = torch.hub.load(repo_or_dir='snakers4/silero-models', model='silero_tts', language=params['language'], speaker=params['model_id']) - model.to(params['device']) - return model -model = load_model() + +"Check if the API is valid and refresh the UI accordingly." +def check_valid_api(): + + global user, user_info, params + + user = ElevenLabsUser(params['api_key']) + user_info = user._get_subscription_data() + print('checking api') + if params['activate'] == False: + return gr.update(value='Disconnected') + elif user_info is None: + print('Incorrect API Key') + return gr.update(value='Disconnected') + else: + print('Got an API Key!') + return gr.update(value='Connected') + +"Once the API is verified, get the available voices and update the dropdown list" +def refresh_voices(): + + global user, user_info + + your_voices = [None] + if user_info is not None: + for voice in user.get_available_voices(): + your_voices.append(voice.initialName) + return gr.Dropdown.update(choices=your_voices) + else: + return def remove_surrounded_chars(string): new_string = "" @@ -46,17 +75,12 @@ def output_modifier(string): """ This function is applied to the model outputs. """ - - global wav_idx, model, current_params - - for i in params: - if params[i] != current_params[i]: - model = load_model() - current_params = params.copy() - break - + global params, wav_idx, user, user_info + if params['activate'] == False: return string + elif user_info == None: + return string string = remove_surrounded_chars(string) string = string.replace('"', '') @@ -66,29 +90,31 @@ def output_modifier(string): if string == '': string = 'empty reply, try regenerating' + + output_file = Path('extensions/elevenlabs_tts/outputs/{}.wav'.format(wav_idx)) + voice = user.get_voices_by_name(params['selected_voice'])[0] + audio_data = voice.generate_audio_bytes(string) + save_bytes_to_path("extensions/elevenlabs_tts/outputs/{}.wav".format(wav_idx), audio_data) - output_file = Path(f'extensions/silero_tts/outputs/{wav_idx:06d}.wav') - audio = model.save_wav(text=string, speaker=params['speaker'], sample_rate=int(params['sample_rate']), audio_path=str(output_file)) string = f'' wav_idx += 1 - return string -def bot_prefix_modifier(string): - """ - This function is only applied in chat mode. It modifies - the prefix text for the Bot and can be used to bias its - behavior. - """ - - return string def ui(): # Gradio elements - activate = gr.Checkbox(value=params['activate'], label='Activate TTS') - voice = gr.Dropdown(value=params['speaker'], choices=voices_by_gender, label='TTS voice') - + with gr.Row(): + activate = gr.Checkbox(value=params['activate'], label='Activate TTS') + connection_status = gr.Textbox(value='Disconnected', label='Connection Status') + voice = gr.Dropdown(value=params['selected_voice'], choices=initial_voice, label='TTS Voice') + with gr.Row(): + api_key = gr.Textbox(placeholder="Enter your API key.", label='API Key') + connect = gr.Button(value='Connect') # Event functions to update the parameters in the backend - activate.change(lambda x: params.update({"activate": x}), activate, None) - voice.change(lambda x: params.update({"speaker": x}), voice, None) + activate.change(lambda x: params.update({'activate': x}), activate, None) + voice.change(lambda x: params.update({'selected_voice': x}), voice, None) + api_key.change(lambda x: params.update({'api_key': x}), api_key, None) + connect.click(check_valid_api, [], connection_status) + connect.click(refresh_voices, [], voice) + From 49ae183ac9662f0062ceb5193d61e6bf1373688f Mon Sep 17 00:00:00 2001 From: oobabooga <112222186+oobabooga@users.noreply.github.com> Date: Mon, 6 Mar 2023 19:28:53 -0300 Subject: [PATCH 2/3] Move new extension to a separate file --- .gitignore | 1 + .../outputs/outputs-will-be-saved-here.txt | 0 extensions/elevenlabs/requirements.txt | 6 + extensions/elevenlabs/script.py | 120 ++++++++++++++++++ extensions/silero_tts/script.py | 102 ++++++--------- 5 files changed, 165 insertions(+), 64 deletions(-) create mode 100644 extensions/elevenlabs/outputs/outputs-will-be-saved-here.txt create mode 100644 extensions/elevenlabs/requirements.txt create mode 100644 extensions/elevenlabs/script.py diff --git a/.gitignore b/.gitignore index 6f4c5ba3..b37a7601 100644 --- a/.gitignore +++ b/.gitignore @@ -1,6 +1,7 @@ cache/* characters/* extensions/silero_tts/outputs/* +extensions/elevenlabs/outputs/* logs/* models/* softprompts/* diff --git a/extensions/elevenlabs/outputs/outputs-will-be-saved-here.txt b/extensions/elevenlabs/outputs/outputs-will-be-saved-here.txt new file mode 100644 index 00000000..e69de29b diff --git a/extensions/elevenlabs/requirements.txt b/extensions/elevenlabs/requirements.txt new file mode 100644 index 00000000..f2f0bff5 --- /dev/null +++ b/extensions/elevenlabs/requirements.txt @@ -0,0 +1,6 @@ +ipython +omegaconf +pydub +PyYAML +torch +torchaudio diff --git a/extensions/elevenlabs/script.py b/extensions/elevenlabs/script.py new file mode 100644 index 00000000..e088ec8e --- /dev/null +++ b/extensions/elevenlabs/script.py @@ -0,0 +1,120 @@ +import asyncio +from pathlib import Path + +import gradio as gr +import torch +import io +import json +import os + +import requests + +from elevenlabslib.helpers import * +from elevenlabslib import * + +params = { + 'activate': True, + 'api_key': '12345', + 'selected_voice': 'None', +} +initial_voice = ['None'] +wav_idx = 0 +user = ElevenLabsUser(params['api_key']) +user_info = None + + +"Check if the API is valid and refresh the UI accordingly." +def check_valid_api(): + + global user, user_info, params + + user = ElevenLabsUser(params['api_key']) + user_info = user._get_subscription_data() + print('checking api') + if params['activate'] == False: + return gr.update(value='Disconnected') + elif user_info is None: + print('Incorrect API Key') + return gr.update(value='Disconnected') + else: + print('Got an API Key!') + return gr.update(value='Connected') + +"Once the API is verified, get the available voices and update the dropdown list" +def refresh_voices(): + + global user, user_info + + your_voices = [None] + if user_info is not None: + for voice in user.get_available_voices(): + your_voices.append(voice.initialName) + return gr.Dropdown.update(choices=your_voices) + else: + return + +def remove_surrounded_chars(string): + new_string = "" + in_star = False + for char in string: + if char == '*': + in_star = not in_star + elif not in_star: + new_string += char + return new_string + +def input_modifier(string): + """ + This function is applied to your text inputs before + they are fed into the model. + """ + + return string + +def output_modifier(string): + """ + This function is applied to the model outputs. + """ + global params, wav_idx, user, user_info + + if params['activate'] == False: + return string + elif user_info == None: + return string + + string = remove_surrounded_chars(string) + string = string.replace('"', '') + string = string.replace('“', '') + string = string.replace('\n', ' ') + string = string.strip() + + if string == '': + string = 'empty reply, try regenerating' + + output_file = Path('extensions/elevenlabs_tts/outputs/{}.wav'.format(wav_idx)) + voice = user.get_voices_by_name(params['selected_voice'])[0] + audio_data = voice.generate_audio_bytes(string) + save_bytes_to_path("extensions/elevenlabs_tts/outputs/{}.wav".format(wav_idx), audio_data) + + + string = f'' + wav_idx += 1 + return string + + +def ui(): + # Gradio elements + with gr.Row(): + activate = gr.Checkbox(value=params['activate'], label='Activate TTS') + connection_status = gr.Textbox(value='Disconnected', label='Connection Status') + voice = gr.Dropdown(value=params['selected_voice'], choices=initial_voice, label='TTS Voice') + with gr.Row(): + api_key = gr.Textbox(placeholder="Enter your API key.", label='API Key') + connect = gr.Button(value='Connect') + # Event functions to update the parameters in the backend + activate.change(lambda x: params.update({'activate': x}), activate, None) + voice.change(lambda x: params.update({'selected_voice': x}), voice, None) + api_key.change(lambda x: params.update({'api_key': x}), api_key, None) + connect.click(check_valid_api, [], connection_status) + connect.click(refresh_voices, [], voice) + diff --git a/extensions/silero_tts/script.py b/extensions/silero_tts/script.py index e088ec8e..f697d0e2 100644 --- a/extensions/silero_tts/script.py +++ b/extensions/silero_tts/script.py @@ -3,55 +3,26 @@ from pathlib import Path import gradio as gr import torch -import io -import json -import os -import requests - -from elevenlabslib.helpers import * -from elevenlabslib import * +torch._C._jit_set_profiling_mode(False) params = { 'activate': True, - 'api_key': '12345', - 'selected_voice': 'None', + 'speaker': 'en_56', + 'language': 'en', + 'model_id': 'v3_en', + 'sample_rate': 48000, + 'device': 'cpu', } -initial_voice = ['None'] +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 -user = ElevenLabsUser(params['api_key']) -user_info = None - -"Check if the API is valid and refresh the UI accordingly." -def check_valid_api(): - - global user, user_info, params - - user = ElevenLabsUser(params['api_key']) - user_info = user._get_subscription_data() - print('checking api') - if params['activate'] == False: - return gr.update(value='Disconnected') - elif user_info is None: - print('Incorrect API Key') - return gr.update(value='Disconnected') - else: - print('Got an API Key!') - return gr.update(value='Connected') - -"Once the API is verified, get the available voices and update the dropdown list" -def refresh_voices(): - - global user, user_info - - your_voices = [None] - if user_info is not None: - for voice in user.get_available_voices(): - your_voices.append(voice.initialName) - return gr.Dropdown.update(choices=your_voices) - else: - return +def load_model(): + model, example_text = torch.hub.load(repo_or_dir='snakers4/silero-models', model='silero_tts', language=params['language'], speaker=params['model_id']) + model.to(params['device']) + return model +model = load_model() def remove_surrounded_chars(string): new_string = "" @@ -75,12 +46,17 @@ def output_modifier(string): """ This function is applied to the model outputs. """ - global params, wav_idx, user, user_info - + + global wav_idx, model, current_params + + for i in params: + if params[i] != current_params[i]: + model = load_model() + current_params = params.copy() + break + if params['activate'] == False: return string - elif user_info == None: - return string string = remove_surrounded_chars(string) string = string.replace('"', '') @@ -90,31 +66,29 @@ def output_modifier(string): if string == '': string = 'empty reply, try regenerating' - - output_file = Path('extensions/elevenlabs_tts/outputs/{}.wav'.format(wav_idx)) - voice = user.get_voices_by_name(params['selected_voice'])[0] - audio_data = voice.generate_audio_bytes(string) - save_bytes_to_path("extensions/elevenlabs_tts/outputs/{}.wav".format(wav_idx), audio_data) + output_file = Path(f'extensions/silero_tts/outputs/{wav_idx:06d}.wav') + audio = model.save_wav(text=string, speaker=params['speaker'], sample_rate=int(params['sample_rate']), audio_path=str(output_file)) string = f'' wav_idx += 1 + return string +def bot_prefix_modifier(string): + """ + This function is only applied in chat mode. It modifies + the prefix text for the Bot and can be used to bias its + behavior. + """ + + return string def ui(): # Gradio elements - with gr.Row(): - activate = gr.Checkbox(value=params['activate'], label='Activate TTS') - connection_status = gr.Textbox(value='Disconnected', label='Connection Status') - voice = gr.Dropdown(value=params['selected_voice'], choices=initial_voice, label='TTS Voice') - with gr.Row(): - api_key = gr.Textbox(placeholder="Enter your API key.", label='API Key') - connect = gr.Button(value='Connect') - # Event functions to update the parameters in the backend - activate.change(lambda x: params.update({'activate': x}), activate, None) - voice.change(lambda x: params.update({'selected_voice': x}), voice, None) - api_key.change(lambda x: params.update({'api_key': x}), api_key, None) - connect.click(check_valid_api, [], connection_status) - connect.click(refresh_voices, [], voice) + activate = gr.Checkbox(value=params['activate'], label='Activate TTS') + voice = gr.Dropdown(value=params['speaker'], choices=voices_by_gender, label='TTS voice') + # Event functions to update the parameters in the backend + activate.change(lambda x: params.update({"activate": x}), activate, None) + voice.change(lambda x: params.update({"speaker": x}), voice, None) From 944fdc03b231492959d633088b8ca0b7ac6b3200 Mon Sep 17 00:00:00 2001 From: oobabooga <112222186+oobabooga@users.noreply.github.com> Date: Mon, 6 Mar 2023 19:38:36 -0300 Subject: [PATCH 3/3] Rename the folder --- .gitignore | 2 +- extensions/elevenlabs/requirements.txt | 6 ------ .../outputs/outputs-will-be-saved-here.txt | 0 extensions/elevenlabs_tts/requirements.txt | 3 +++ extensions/{elevenlabs => elevenlabs_tts}/script.py | 10 ++++------ 5 files changed, 8 insertions(+), 13 deletions(-) delete mode 100644 extensions/elevenlabs/requirements.txt rename extensions/{elevenlabs => elevenlabs_tts}/outputs/outputs-will-be-saved-here.txt (100%) create mode 100644 extensions/elevenlabs_tts/requirements.txt rename extensions/{elevenlabs => elevenlabs_tts}/script.py (99%) diff --git a/.gitignore b/.gitignore index b37a7601..1b7f0fb8 100644 --- a/.gitignore +++ b/.gitignore @@ -1,7 +1,7 @@ cache/* characters/* extensions/silero_tts/outputs/* -extensions/elevenlabs/outputs/* +extensions/elevenlabs_tts/outputs/* logs/* models/* softprompts/* diff --git a/extensions/elevenlabs/requirements.txt b/extensions/elevenlabs/requirements.txt deleted file mode 100644 index f2f0bff5..00000000 --- a/extensions/elevenlabs/requirements.txt +++ /dev/null @@ -1,6 +0,0 @@ -ipython -omegaconf -pydub -PyYAML -torch -torchaudio diff --git a/extensions/elevenlabs/outputs/outputs-will-be-saved-here.txt b/extensions/elevenlabs_tts/outputs/outputs-will-be-saved-here.txt similarity index 100% rename from extensions/elevenlabs/outputs/outputs-will-be-saved-here.txt rename to extensions/elevenlabs_tts/outputs/outputs-will-be-saved-here.txt diff --git a/extensions/elevenlabs_tts/requirements.txt b/extensions/elevenlabs_tts/requirements.txt new file mode 100644 index 00000000..8ec07a8a --- /dev/null +++ b/extensions/elevenlabs_tts/requirements.txt @@ -0,0 +1,3 @@ +elevenlabslib +soundfile +sounddevice diff --git a/extensions/elevenlabs/script.py b/extensions/elevenlabs_tts/script.py similarity index 99% rename from extensions/elevenlabs/script.py rename to extensions/elevenlabs_tts/script.py index e088ec8e..ae4fe019 100644 --- a/extensions/elevenlabs/script.py +++ b/extensions/elevenlabs_tts/script.py @@ -1,16 +1,14 @@ import asyncio -from pathlib import Path - -import gradio as gr -import torch import io import json import os +from pathlib import Path +import gradio as gr import requests - -from elevenlabslib.helpers import * +import torch from elevenlabslib import * +from elevenlabslib.helpers import * params = { 'activate': True,