xinference 0.14.4.post1__py3-none-any.whl → 0.15.1__py3-none-any.whl

xinference/thirdparty/fish_speech/fish_speech/text/chn_text_norm/README.md

@@ -0,0 +1,36 @@
+# This account is no longer in use, see [Atomicoo](https://github.com/atomicoo) for my latest works.
+
+# Chn Text Norm
+
+this is a repository for chinese text normalization (no longer maintained).
+
+## Quick Start ##
+
+### Git Clone Repo ###
+
+git clone this repo to the root directory of your project which need to use it.
+
+    cd /path/to/proj
+    git clone https://github.com/Joee1995/chn-text-norm.git
+
+after that, your doc tree should be:
+```
+proj                     # root of your project
+|--- chn_text_norm       # this chn-text-norm tool
+     |--- text.py
+     |--- ...
+|--- text_normalize.py   # your text normalization code
+|--- ...
+```
+
+### How to Use ? ###
+
+    # text_normalize.py
+    from chn_text_norm.text import *
+    
+    raw_text = 'your raw text'
+    text = Text(raw_text=raw_text).normalize()
+
+### How to add quantums ###
+
+打开test.py，然后你就知道怎么做了。

xinference/thirdparty/fish_speech/fish_speech/text/clean.py

@@ -1,61 +1,24 @@
-import itertools
 import re
 
-LANGUAGE_UNICODE_RANGE_MAP = {
-    "ZH": [(0x4E00, 0x9FFF)],
-    "JP": [(0x4E00, 0x9FFF), (0x3040, 0x309F), (0x30A0, 0x30FF), (0x31F0, 0x31FF)],
-    "EN": [(0x0000, 0x007F)],
-}
-
 SYMBOLS_MAPPING = {
-    "：": ",",
-    "；": ",",
-    "，": ",",
-    "。": ".",
-    "！": "!",
-    "？": "?",
-    "\n": ".",
-    "·": ",",
-    "、": ",",
-    "...": "…",
     "“": "'",
     "”": "'",
     "‘": "'",
     "’": "'",
-    "（": "'",
-    "）": "'",
-    "(": "'",
-    ")": "'",
-    "《": "'",
-    "》": "'",
-    "【": "'",
-    "】": "'",
-    "[": "'",
-    "]": "'",
-    "—": "-",
-    "～": "-",
-    "~": "-",
-    "・": "-",
-    "「": "'",
-    "」": "'",
-    ";": ",",
-    ":": ",",
+    "【": "",
+    "】": "",
+    "[": "",
+    "]": "",
+    "（": "",
+    "）": "",
+    "(": "",
+    ")": "",
+    "・": "·",
 }
 
 REPLACE_SYMBOL_REGEX = re.compile(
     "|".join(re.escape(p) for p in SYMBOLS_MAPPING.keys())
 )
-ALL_KNOWN_UTF8_RANGE = list(
-    itertools.chain.from_iterable(LANGUAGE_UNICODE_RANGE_MAP.values())
-)
-REMOVE_UNKNOWN_SYMBOL_REGEX = re.compile(
-    "[^"
-    + "".join(
-        f"{re.escape(chr(start))}-{re.escape(chr(end))}"
-        for start, end in ALL_KNOWN_UTF8_RANGE
-    )
-    + "]"
-)
 
 
 def clean_text(text):
@@ -64,6 +27,5 @@ def clean_text(text):
 
     # Replace all chinese symbols with their english counterparts
     text = REPLACE_SYMBOL_REGEX.sub(lambda x: SYMBOLS_MAPPING[x.group()], text)
-    text = REMOVE_UNKNOWN_SYMBOL_REGEX.sub("", text)
 
     return text

xinference/thirdparty/fish_speech/fish_speech/text/spliter.py

@@ -71,9 +71,9 @@ def split_text(text, length):
 
     texts = [text]
     texts = map(protect_float, texts)
-    texts = break_text(texts, length, {".", "!", "?"})
+    texts = break_text(texts, length, {".", "!", "?", "。", "！", "？"})
     texts = map(unprotect_float, texts)
-    texts = break_text(texts, length, {","})
+    texts = break_text(texts, length, {",", "，"})
     texts = break_text(texts, length, {" "})
     texts = list(break_text_by_length(texts, length))
 

xinference/thirdparty/fish_speech/fish_speech/train.py

@@ -1,4 +1,6 @@
 import os
+
+os.environ["USE_LIBUV"] = "0"
 import sys
 from typing import Optional
 

xinference/thirdparty/fish_speech/fish_speech/webui/css/style.css

@@ -0,0 +1,161 @@
+:root {
+  --my-200: #80eeee;
+  --my-50: #ecfdf5;
+  --water-width: 300px;
+  --water-heigh: 300px;
+}
+
+
+/* general styled components */
+.tools {
+  align-items: center;
+  justify-content: center;
+}
+
+.gradio-button {
+    max-width: 2.2em;
+    min-width: 2.2em !important;
+    height: 2.4em;
+    align-self: end;
+    line-height: 1em;
+    border-radius: 0.5em;
+
+}
+
+.gradio-button.secondary-down, .gradio-button.secondary-down:hover{
+    box-shadow: 1px 1px 1px rgba(0,0,0,0.25) inset, 0px 0px 3px rgba(0,0,0,0.15) inset;
+}
+
+/* replace original footer with ours */
+a{
+    font-weight: bold;
+    cursor: pointer;
+    color: #030C14 !important;
+}
+
+footer {
+    display: none !important;
+}
+
+#footer{
+    text-align: center;
+}
+
+#footer div{
+    display: inline-block;
+}
+
+#footer .versions{
+    font-size: 85%;
+    opacity: 0.85;
+}
+
+/*@keyframes moveBackground {*/
+/*  0% {*/
+/*    background-position: 0 0;*/
+/*  }*/
+/*  100% {*/
+/*    background-position: -100px 100px;*/
+/*  }*/
+/*}*/
+@keyframes moveJellyBackground {
+  0% {
+    background-position: 0% 50%;
+  }
+  50% {
+    background-position: 100% 50%;
+  }
+  100% {
+    background-position: 0% 50%;
+  }
+}
+
+.gradio-container {
+  position: absolute;
+  z-index: 10;
+}
+
+
+.quan {
+  position: absolute;
+  bottom: 0;
+  width: var(--water-width);
+  height: var(--water-heigh);
+  border-radius: 0;
+  /*border: 3px solid rgb(246, 247, 248);*/
+  /*box-shadow: 0 0 0 3px rgb(41, 134, 196);*/
+  z-index: 0;
+
+}
+
+.quan:last-child {
+  margin-right: 0;
+}
+
+.shui {
+  position: absolute;
+  top: 0;
+  left: 0;
+  width: 100%;
+  height: 100%;
+  background-color: rgb(23, 106, 201);
+  border-radius: 0;
+  overflow: hidden;
+  z-index: 0;
+}
+
+.shui::after {
+
+  content: '';
+  position: absolute;
+  top: 20%;
+  left: 50%;
+  width: 150%;
+  height: 150%;
+  border-radius: 40%;
+  background-image: radial-gradient(circle at 0% 50%, #dcfcf1, var(--my-50) 50%);
+  animation: shi 5s linear infinite;
+}
+
+@keyframes shi {
+  0% {
+    transform: translate(-50%, -65%) rotate(0deg);
+  }
+  100% {
+    transform: translate(-50%, -65%) rotate(360deg);
+  }
+}
+
+.shui::before {
+  content: '';
+  position: absolute;
+  top: 20%;
+  left: 50%;
+  width: 150%;
+  height: 150%;
+  border-radius: 42%;
+  background-color: rgb(240, 228, 228, 0.2);
+  animation: xu 7s linear infinite;
+}
+
+@keyframes xu {
+  0% {
+    transform: translate(-50%, -60%) rotate(0deg);
+  }
+  100% {
+    transform: translate(-50%, -60%) rotate(360deg);
+  }
+}
+
+fieldset.data_src div.wrap label {
+  background: #f8bffee0 !important;
+}
+
+.scrollable-component {
+  max-height: 100px;
+  overflow-y: auto;
+}
+
+#file_accordion {
+  max-height: 220px !important;
+}

xinference/thirdparty/fish_speech/fish_speech/webui/html/footer.html

@@ -0,0 +1,11 @@
+<div style="color: rgba(25,255,205,0.7) !important;">
+        <a href="{api_docs}">API</a>
+         • 
+        <a href="https://github.com/fishaudio/fish-speech">Github</a>
+         • 
+        <a href="https://gradio.app">Gradio</a>
+</div>
+<br />
+<div class="versions" style="color: rgba(25,255,205,0.7) !important;">
+{versions}
+</div>

xinference/thirdparty/fish_speech/fish_speech/webui/js/animate.js

@@ -0,0 +1,69 @@
+
+function createGradioAnimation() {
+    const params = new URLSearchParams(window.location.search);
+    if (!params.has('__theme')) {
+        params.set('__theme', 'light');
+        window.location.search = params.toString();
+    }
+
+    var gradioApp = document.querySelector('gradio-app');
+    if (gradioApp) {
+
+        document.documentElement.style.setProperty('--my-200', '#80eeee');
+        document.documentElement.style.setProperty('--my-50', '#ecfdf5');
+
+        // gradioApp.style.position = 'relative';
+        // gradioApp.style.backgroundSize = '200% 200%';
+        // gradioApp.style.animation = 'moveJellyBackground 10s ease infinite';
+        // gradioApp.style.backgroundImage = 'radial-gradient(circle at 0% 50%, var(--my-200), var(--my-50) 50%)';
+        // gradioApp.style.display = 'flex';
+        // gradioApp.style.justifyContent = 'flex-start';
+        // gradioApp.style.flexWrap = 'nowrap';
+        // gradioApp.style.overflowX = 'auto';
+
+        // for (let i = 0; i < 6; i++) {
+        //     var quan = document.createElement('div');
+        //     quan.className = 'quan';
+        //     gradioApp.insertBefore(quan, gradioApp.firstChild);
+        //     quan.id = 'quan' + i.toString();
+        //     quan.style.left = 'calc(var(--water-width) * ' + i.toString() + ')';
+        //     var quanContainer = document.querySelector('.quan');
+        //     if (quanContainer) {
+        //         var shui = document.createElement('div');
+        //         shui.className = 'shui';
+        //         quanContainer.insertBefore(shui, quanContainer.firstChild)
+        //     }
+        // }
+    }
+
+    var container = document.createElement('div');
+    container.id = 'gradio-animation';
+    container.style.fontSize = '2em';
+    container.style.fontFamily = 'Maiandra GD, ui-monospace, monospace';
+    container.style.fontWeight = 'bold';
+    container.style.textAlign = 'center';
+    container.style.marginBottom = '20px';
+
+    var text = 'Welcome to Fish-Speech!';
+    for (var i = 0; i < text.length; i++) {
+        (function(i){
+            setTimeout(function(){
+                var letter = document.createElement('span');
+                letter.style.opacity = '0';
+                letter.style.transition = 'opacity 0.5s';
+                letter.innerText = text[i];
+
+                container.appendChild(letter);
+
+                setTimeout(function() {
+                    letter.style.opacity = '1';
+                }, 50);
+            }, i * 200);
+        })(i);
+    }
+
+    var gradioContainer = document.querySelector('.gradio-container');
+    gradioContainer.insertBefore(container, gradioContainer.firstChild);
+
+    return 'Animation created';
+}

xinference/thirdparty/fish_speech/fish_speech/webui/manage.py

@@ -1,9 +1,11 @@
 from __future__ import annotations
 
+import os
+
+os.environ["USE_LIBUV"] = "0"
 import datetime
 import html
 import json
-import os
 import platform
 import shutil
 import signal
@@ -469,7 +471,7 @@ def train_process(
                 "--config-name",
                 "firefly_gan_vq",
                 "--checkpoint-path",
-                "checkpoints/fish-speech-1.2-sft/firefly-gan-vq-fsq-4x1024-42hz-generator.pth",
+                "checkpoints/fish-speech-1.4/firefly-gan-vq-fsq-8x1024-21hz-generator.pth",
             ]
         )
 
@@ -485,7 +487,7 @@ def train_process(
                 "16",
             ]
         )
-        ckpt_path = "checkpoints/fish-speech-1.2-sft/model.pth"
+        ckpt_path = "checkpoints/fish-speech-1.4/model.pth"
         lora_prefix = "lora_" if llama_use_lora else ""
         llama_name = lora_prefix + "text2semantic_" + new_project
         latest = next(
@@ -862,7 +864,7 @@ with gr.Blocks(
                                     minimum=1,
                                     maximum=32,
                                     step=1,
-                                    value=4,
+                                    value=2,
                                 )
                                 llama_data_max_length_slider = gr.Slider(
                                     label=i18n("Maximum Length per Sample"),
@@ -870,7 +872,7 @@ with gr.Blocks(
                                     minimum=1024,
                                     maximum=4096,
                                     step=128,
-                                    value=1024,
+                                    value=2048,
                                 )
                             with gr.Row(equal_height=False):
                                 llama_precision_dropdown = gr.Dropdown(
@@ -925,9 +927,9 @@ with gr.Blocks(
                                         "Type the path or select from the dropdown"
                                     ),
                                     choices=[
-                                        "checkpoints/fish-speech-1.2-sft/model.pth",
+                                        "checkpoints/fish-speech-1.4/model.pth",
                                     ],
-                                    value="checkpoints/fish-speech-1.2-sft/model.pth",
+                                    value="checkpoints/fish-speech-1.4/model.pth",
                                     allow_custom_value=True,
                                     interactive=True,
                                 )
@@ -979,7 +981,7 @@ with gr.Blocks(
                                         "Type the path or select from the dropdown"
                                     ),
                                     choices=list_llama_models(),
-                                    value="checkpoints/fish-speech-1.2-sft",
+                                    value="checkpoints/fish-speech-1.4",
                                     allow_custom_value=True,
                                     interactive=True,
                                 )
@@ -1042,7 +1044,7 @@ with gr.Blocks(
                                         "Type the path or select from the dropdown"
                                     ),
                                     choices=list_decoder_models(),
-                                    value="checkpoints/fish-speech-1.2-sft/firefly-gan-vq-fsq-4x1024-42hz-generator.pth",
+                                    value="checkpoints/fish-speech-1.4/firefly-gan-vq-fsq-8x1024-21hz-generator.pth",
                                     allow_custom_value=True,
                                 )
                                 infer_decoder_config = gr.Dropdown(
@@ -1060,7 +1062,7 @@ with gr.Blocks(
                                     info=i18n(
                                         "Type the path or select from the dropdown"
                                     ),
-                                    value="checkpoints/fish-speech-1.2-sft",
+                                    value="checkpoints/fish-speech-1.4",
                                     choices=list_llama_models(),
                                     allow_custom_value=True,
                                 )