232 lines
9.5 KiB
Python
232 lines
9.5 KiB
Python
from typing import Callable
|
|
|
|
import gradio as gr
|
|
|
|
from fish_speech.i18n import i18n
|
|
from tools.inference_engine.utils import normalize_text
|
|
from tools.webui.variables import HEADER_MD, TEXTBOX_PLACEHOLDER
|
|
|
|
|
|
def build_app(inference_fct: Callable, theme: str = "light") -> gr.Blocks:
|
|
with gr.Blocks(theme=gr.themes.Base()) as app:
|
|
gr.Markdown(HEADER_MD)
|
|
|
|
# Use light theme by default
|
|
app.load(
|
|
None,
|
|
None,
|
|
js="() => {const params = new URLSearchParams(window.location.search);if (!params.has('__theme')) {params.set('__theme', '%s');window.location.search = params.toString();}}"
|
|
% theme,
|
|
)
|
|
|
|
# Inference
|
|
with gr.Row():
|
|
with gr.Column(scale=3):
|
|
text = gr.Textbox(
|
|
label=i18n("Input Text"), placeholder=TEXTBOX_PLACEHOLDER, lines=10
|
|
)
|
|
refined_text = gr.Textbox(
|
|
label=i18n("Realtime Transform Text"),
|
|
placeholder=i18n(
|
|
"Normalization Result Preview (Currently Only Chinese)"
|
|
),
|
|
lines=5,
|
|
interactive=False,
|
|
)
|
|
|
|
with gr.Row():
|
|
normalize = gr.Checkbox(
|
|
label=i18n("Text Normalization"),
|
|
value=False,
|
|
)
|
|
|
|
with gr.Row():
|
|
with gr.Column():
|
|
with gr.Tab(label=i18n("Reference Audio")):
|
|
with gr.Row():
|
|
gr.Markdown(
|
|
i18n(
|
|
"5 to 10 seconds of reference audio, useful for specifying speaker."
|
|
)
|
|
)
|
|
with gr.Row():
|
|
reference_id = gr.Textbox(
|
|
label=i18n("Reference ID"),
|
|
placeholder=i18n("Leave empty to use uploaded references"),
|
|
)
|
|
|
|
with gr.Row():
|
|
use_memory_cache = gr.Radio(
|
|
label=i18n("Use Memory Cache"),
|
|
choices=[i18n("on"), i18n("off")],
|
|
value=i18n("on"),
|
|
)
|
|
|
|
with gr.Row():
|
|
from pathlib import Path
|
|
initial_audio_files = [str(p) for p in Path("references").glob("**/*") if p.suffix.lower() in [".wav", ".mp3", ".flac"]]
|
|
example_audio_dropdown = gr.Dropdown(
|
|
label=i18n("Select Example Audio"),
|
|
choices=initial_audio_files + [""],
|
|
value="",
|
|
interactive=True,
|
|
allow_custom_value=True,
|
|
)
|
|
refresh_button = gr.Button(
|
|
value="\U0001F504 刷新",
|
|
)
|
|
|
|
with gr.Row():
|
|
reference_audio = gr.Audio(
|
|
label=i18n("Reference Audio"),
|
|
type="filepath",
|
|
)
|
|
with gr.Row():
|
|
reference_text = gr.Textbox(
|
|
label=i18n("Reference Text"),
|
|
lines=1,
|
|
placeholder="在一无所知中，梦里的一天结束了，一个新的「轮回」便会开始。",
|
|
value="",
|
|
)
|
|
|
|
with gr.Tab(label=i18n("Advanced Config")):
|
|
with gr.Row():
|
|
chunk_length = gr.Slider(
|
|
label=i18n("Iterative Prompt Length, 0 means off"),
|
|
minimum=0,
|
|
maximum=300,
|
|
value=200,
|
|
step=8,
|
|
)
|
|
|
|
max_new_tokens = gr.Slider(
|
|
label=i18n(
|
|
"Maximum tokens per batch, 0 means no limit"
|
|
),
|
|
minimum=0,
|
|
maximum=2048,
|
|
value=0,
|
|
step=8,
|
|
)
|
|
|
|
with gr.Row():
|
|
top_p = gr.Slider(
|
|
label="Top-P",
|
|
minimum=0.6,
|
|
maximum=0.9,
|
|
value=0.7,
|
|
step=0.01,
|
|
)
|
|
|
|
repetition_penalty = gr.Slider(
|
|
label=i18n("Repetition Penalty"),
|
|
minimum=1,
|
|
maximum=1.5,
|
|
value=1.2,
|
|
step=0.01,
|
|
)
|
|
|
|
with gr.Row():
|
|
temperature = gr.Slider(
|
|
label=i18n("Temperature"),
|
|
minimum=0.6,
|
|
maximum=0.9,
|
|
value=0.7,
|
|
step=0.01,
|
|
)
|
|
seed = gr.Number(
|
|
label=i18n("Seed"),
|
|
info=i18n("0 means randomized inference, otherwise deterministic"),
|
|
value=0,
|
|
)
|
|
|
|
with gr.Column(scale=3):
|
|
with gr.Row():
|
|
error = gr.HTML(
|
|
label=i18n("Error Message"),
|
|
visible=True,
|
|
)
|
|
with gr.Row():
|
|
audio = gr.Audio(
|
|
label=i18n("Generated Audio"),
|
|
type="numpy",
|
|
interactive=False,
|
|
visible=True,
|
|
)
|
|
|
|
with gr.Row():
|
|
used_seed = gr.Textbox(
|
|
label=i18n("Used Seed"),
|
|
value="",
|
|
interactive=False,
|
|
)
|
|
|
|
with gr.Row():
|
|
with gr.Column(scale=3):
|
|
generate = gr.Button(
|
|
value="\U0001F3A7 " + i18n("Generate"),
|
|
variant="primary",
|
|
)
|
|
|
|
text.input(fn=normalize_text, inputs=[text, normalize], outputs=[refined_text])
|
|
|
|
# Refresh audio files list and update dropdown
|
|
def refresh_audio_files():
|
|
audio_files = [str(p) for p in Path("references").glob("**/*") if p.suffix.lower() in [".wav", ".mp3", ".flac"]]
|
|
return gr.Dropdown(choices=audio_files + [""])
|
|
|
|
refresh_button.click(
|
|
fn=refresh_audio_files,
|
|
inputs=[],
|
|
outputs=[example_audio_dropdown],
|
|
)
|
|
|
|
# Update reference audio and text when dropdown changes
|
|
def update_reference_audio_and_text(audio_path):
|
|
if audio_path and audio_path != "":
|
|
# Extract filename without extension as reference text
|
|
reference_text = Path(audio_path).stem
|
|
return audio_path, reference_text
|
|
return None, ""
|
|
|
|
example_audio_dropdown.change(
|
|
fn=update_reference_audio_and_text,
|
|
inputs=[example_audio_dropdown],
|
|
outputs=[reference_audio, reference_text],
|
|
)
|
|
|
|
# Submit
|
|
def preprocess_inputs(refined_text, normalize, reference_id, reference_audio, reference_text,
|
|
max_new_tokens, chunk_length, top_p, repetition_penalty, temperature,
|
|
seed, use_memory_cache):
|
|
# Convert localized value back to English
|
|
if use_memory_cache == i18n("on"):
|
|
use_memory_cache = "on"
|
|
else:
|
|
use_memory_cache = "off"
|
|
return [refined_text, normalize, reference_id, reference_audio, reference_text,
|
|
max_new_tokens, chunk_length, top_p, repetition_penalty, temperature,
|
|
seed, use_memory_cache]
|
|
|
|
generate.click(
|
|
lambda *args: inference_fct(*preprocess_inputs(*args)),
|
|
[
|
|
refined_text,
|
|
normalize,
|
|
reference_id,
|
|
reference_audio,
|
|
reference_text,
|
|
max_new_tokens,
|
|
chunk_length,
|
|
top_p,
|
|
repetition_penalty,
|
|
temperature,
|
|
seed,
|
|
use_memory_cache,
|
|
],
|
|
[audio, error, used_seed],
|
|
concurrency_limit=1,
|
|
)
|
|
|
|
return app
|