Paillat-dev/viralfactory
1
0
Fork 0
mirror of https://github.com/Paillat-dev/viralfactory.git synced 2026-01-02 01:06:19 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix import, fix option indices in init, formatting

Browse Source
This commit is contained in:
Paillat
2024-02-15 17:52:19 +01:00
parent c57817df85
commit 67237a5ecf
1 changed files with 31 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

52
src/engines/TTSEngine/CoquiTTSEngine.py
View File

@@ -1,6 +1,6 @@
import gradio as gr import gradio as gr
import TTS from TTS.api import TTS
import os import os
import torch import torch
@@ -95,9 +95,9 @@ class CoquiTTSEngine(BaseTTSEngine):
def __init__(self, options: list): def __init__(self, options: list):
super().__init__() super().__init__()
self.voice = options[0][0] self.voice = options[0]
self.language = options[1][0] self.language = options[1]
self.to_force_duration = options[2][0] self.to_force_duration = options[2]
self.duration = options[3] self.duration = options[3]
os.environ["COQUI_TOS_AGREED"] = "1" os.environ["COQUI_TOS_AGREED"] = "1"
@@ -107,20 +107,22 @@ class CoquiTTSEngine(BaseTTSEngine):
self.tts.to(device) self.tts.to(device)
def synthesize(self, text: str, path: str): def synthesize(self, text: str, path: str):
""" """
Synthesizes the given text into speech and saves it to the specified file path. Synthesizes the given text into speech and saves it to the specified file path.
Args: Args:
text (str): The text to synthesize into speech. text (str): The text to synthesize into speech.
path (str): The file path to save the synthesized speech. path (str): The file path to save the synthesized speech.
Returns: Returns:
float: The time taken to synthesize the speech with whispering effect. float: The time taken to synthesize the speech with whispering effect.
""" """
self.tts.tts_to_file(text=text, file_path=path, lang=self.language, speaker=self.voice) self.tts.tts_to_file(
if self.to_force_duration: text=text, file_path=path, lang=self.language, speaker=self.voice
self.force_duration(float(self.duration), path) )
return self.time_with_whisper(path) if self.to_force_duration:
self.force_duration(float(self.duration), path)
return self.time_with_whisper(path)
@classmethod @classmethod
def get_options(cls) -> list: def get_options(cls) -> list:
@@ -138,12 +140,20 @@ class CoquiTTSEngine(BaseTTSEngine):
value=cls.languages[0], value=cls.languages[0],
), ),
] ]
duration_checkbox = gr.Checkbox(label="Force duration", info="Force the duration of the generated audio to be at most the specified value", value=False) duration_checkbox = gr.Checkbox(
duration = gr.Number(label="Duration [s]", value=57, step=1, minimum=10, visible=False) label="Force duration",
info="Force the duration of the generated audio to be at most the specified value",
value=False,
)
duration = gr.Number(
label="Duration [s]", value=57, step=1, minimum=10, visible=False
)
duration_switch = lambda x: gr.update(visible=x) duration_switch = lambda x: gr.update(visible=x)
duration_checkbox.change(duration_switch, inputs=[duration_checkbox], outputs=[duration]) duration_checkbox.change(
duration_switch, inputs=[duration_checkbox], outputs=[duration]
)
options.append(duration_checkbox) options.append(duration_checkbox)
options.append(duration) options.append(duration)
return options return options