Upload 6 files

.gitattributes

@@ -1,35 +1,35 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

README.md

@@ -1,14 +1,14 @@
----
-title: Text To Speech TTS
-emoji: 🏢
-colorFrom: pink
-colorTo: pink
-sdk: gradio
-sdk_version: 5.38.2
-app_file: app.py
-pinned: false
-license: apache-2.0
-short_description: Convert text to Speech
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+---
+title: tts Text To Speech
+emoji: 🌍
+colorFrom: yellow
+colorTo: pink
+sdk: gradio
+sdk_version: 4.44.1
+python_version: 3.8.9
+app_file: app.py
+pinned: false
+license: apache-2.0
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,274 @@
+#!/usr/bin/env python3
+#
+# Copyright      2022-2023  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# References:
+# https://gradio.app/docs/#dropdown
+
+import os
+import time
+import uuid
+from datetime import datetime
+
+import gradio as gr
+import soundfile as sf
+
+from model import get_pretrained_model, language_to_models
+
+
+def MyPrint(s):
+    now = datetime.now()
+    date_time = now.strftime("%Y-%m-%d %H:%M:%S.%f")
+    print(f"{date_time}: {s}")
+
+
+title = "# Next-gen Kaldi: Text-to-speech (TTS)"
+
+description = """
+This space shows how to convert text to speech with Next-gen Kaldi.
+
+It is running on CPU within a docker container provided by Hugging Face.
+
+See more information by visiting the following links:
+
+- <https://github.com/k2-fsa/sherpa-onnx>
+
+If you want to deploy it locally, please see
+<https://k2-fsa.github.io/sherpa/>
+
+If you want to use Android APKs, please see
+<https://k2-fsa.github.io/sherpa/onnx/tts/apk.html>
+
+If you want to use Android text-to-speech engine APKs, please see
+<https://k2-fsa.github.io/sherpa/onnx/tts/apk-engine.html>
+
+If you want to download an all-in-one exe for Windows, please see
+<https://github.com/k2-fsa/sherpa-onnx/releases/tag/tts-models>
+
+"""
+
+# css style is copied from
+# https://huggingface.co/spaces/alphacep/asr/blob/main/app.py#L113
+css = """
+.result {display:flex;flex-direction:column}
+.result_item {padding:15px;margin-bottom:8px;border-radius:15px;width:100%}
+.result_item_success {background-color:mediumaquamarine;color:white;align-self:start}
+.result_item_error {background-color:#ff7070;color:white;align-self:start}
+"""
+
+examples = [
+    [
+        "Chinese (Mandarin, 普通话)",
+        "csukuangfj/matcha-icefall-zh-baker|1 speaker",
+        "某某银行的副行长和一些行政领导表示，他们去过长江和长白山; 经济不断增长。2024年12月31号，拨打110或者18920240511。123456块钱。",
+        0,
+        1.0,
+    ],
+    [
+        "Chinese (Mandarin, 普通话)",
+        "csukuangfj/vits-zh-hf-fanchen-wnj|1 speaker",
+        "在一个阳光明媚的夏天，小马、小羊和小狗它们一块儿在广阔的草地上，嬉戏玩耍，这时小猴来了，还带着它心爱的足球活蹦乱跳地跑前、跑后教小马、小羊、小狗踢足球。",
+        0,
+        1.0,
+    ],
+    [
+        "Chinese (Mandarin, 普通话)",
+        "csukuangfj/vits-zh-hf-fanchen-C|187 speakers",
+        '小米的使命是，始终坚持做"感动人心、价格厚道"的好产品，让全球每个人都能享受科技带来的美好生活。',
+        0,
+        1.0,
+    ],
+    ["Min-nan (闽南话)", "csukuangfj/vits-mms-nan", "ài piaǸ chiah ē iaN̂", 0, 1.0],
+    ["Thai", "csukuangfj/vits-mms-tha", "ฉันรักคุณ", 0, 1.0],
+    [
+        "Chinese (Mandarin, 普通话)",
+        "csukuangfj/sherpa-onnx-vits-zh-ll|5 speakers",
+        "当夜幕降临，星光点点，伴随着微风拂面，我在静谧中感受着时光的流转，思念如涟漪荡漾，梦境如画卷展开，我与自然融为一体，沉静在这片宁静的美丽之中，感受着生命的奇迹与温柔。",
+        2,
+        1.0,
+    ],
+]
+
+
+def update_model_dropdown(language: str):
+    if language in language_to_models:
+        choices = language_to_models[language]
+        return gr.Dropdown(
+            choices=choices,
+            value=choices[0],
+            interactive=True,
+        )
+
+    raise ValueError(f"Unsupported language: {language}")
+
+
+def build_html_output(s: str, style: str = "result_item_success"):
+    return f"""
+    <div class='result'>
+        <div class='result_item {style}'>
+          {s}
+        </div>
+    </div>
+    """
+
+
+def process(language: str, repo_id: str, text: str, sid: str, speed: float):
+    MyPrint(f"Input text: {text}. sid: {sid}, speed: {speed}")
+    sid = int(sid)
+    tts = get_pretrained_model(repo_id, speed)
+
+    start = time.time()
+    audio = tts.generate(text, sid=sid)
+    end = time.time()
+
+    if len(audio.samples) == 0:
+        raise ValueError(
+            "Error in generating audios. Please read previous error messages."
+        )
+
+    duration = len(audio.samples) / audio.sample_rate
+
+    elapsed_seconds = end - start
+    rtf = elapsed_seconds / duration
+
+    info = f"""
+    Wave duration  : {duration:.3f} s <br/>
+    Processing time: {elapsed_seconds:.3f} s <br/>
+    RTF: {elapsed_seconds:.3f}/{duration:.3f} = {rtf:.3f} <br/>
+    """
+
+    MyPrint(info)
+    MyPrint(f"\nrepo_id: {repo_id}\ntext: {text}\nsid: {sid}\nspeed: {speed}")
+
+    filename = str(uuid.uuid4())
+    filename = f"{filename}.wav"
+    sf.write(
+        filename,
+        audio.samples,
+        samplerate=audio.sample_rate,
+        subtype="PCM_16",
+    )
+
+    return filename, build_html_output(info)
+
+
+demo = gr.Blocks(css=css)
+
+
+with demo:
+    gr.Markdown(title)
+    language_choices = list(language_to_models.keys())
+
+    language_radio = gr.Radio(
+        label="Language",
+        choices=language_choices,
+        value=language_choices[0],
+    )
+
+    model_dropdown = gr.Dropdown(
+        choices=language_to_models[language_choices[0]],
+        label="Select a model",
+        value=language_to_models[language_choices[0]][0],
+    )
+
+    language_radio.change(
+        update_model_dropdown,
+        inputs=language_radio,
+        outputs=model_dropdown,
+    )
+
+    with gr.Tabs():
+        with gr.TabItem("Please input your text"):
+            input_text = gr.Textbox(
+                label="Input text",
+                info="Your text",
+                lines=3,
+                placeholder="Please input your text here",
+            )
+
+            input_sid = gr.Textbox(
+                label="Speaker ID",
+                info="Speaker ID",
+                lines=1,
+                max_lines=1,
+                value="0",
+                placeholder="Speaker ID. Valid only for mult-speaker model",
+            )
+
+            input_speed = gr.Slider(
+                minimum=0.1,
+                maximum=10,
+                value=1,
+                step=0.1,
+                label="Speed (larger->faster; smaller->slower)",
+            )
+
+            input_button = gr.Button("Submit")
+
+            output_audio = gr.Audio(label="Output")
+
+            output_info = gr.HTML(label="Info")
+
+            gr.Examples(
+                examples=examples,
+                fn=process,
+                inputs=[
+                    language_radio,
+                    model_dropdown,
+                    input_text,
+                    input_sid,
+                    input_speed,
+                ],
+                outputs=[
+                    output_audio,
+                    output_info,
+                ],
+            )
+
+        input_button.click(
+            process,
+            inputs=[
+                language_radio,
+                model_dropdown,
+                input_text,
+                input_sid,
+                input_speed,
+            ],
+            outputs=[
+                output_audio,
+                output_info,
+            ],
+        )
+
+    gr.Markdown(description)
+
+
+def download_espeak_ng_data():
+    os.system(
+        """
+    cd /tmp
+    wget -qq https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/espeak-ng-data.tar.bz2
+    tar xf espeak-ng-data.tar.bz2
+    """
+    )
+
+
+if __name__ == "__main__":
+    download_espeak_ng_data()
+    formatter = "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
+
+    demo.launch()

model.py

@@ -0,0 +1,1071 @@
+# Copyright      2022-2023  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+from functools import lru_cache
+from pathlib import Path
+
+import sherpa_onnx
+from huggingface_hub import hf_hub_download
+
+
+def get_file(
+    repo_id: str,
+    filename: str,
+    subfolder: str = ".",
+) -> str:
+    model_filename = hf_hub_download(
+        repo_id=repo_id,
+        filename=filename,
+        subfolder=subfolder,
+    )
+    return model_filename
+
+
+@lru_cache(maxsize=10)
+def _get_vits_vctk(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    assert repo_id == "csukuangfj/vits-vctk"
+
+    model = get_file(
+        repo_id=repo_id,
+        filename="vits-vctk.onnx",
+        subfolder=".",
+    )
+
+    lexicon = get_file(
+        repo_id=repo_id,
+        filename="lexicon.txt",
+        subfolder=".",
+    )
+
+    tokens = get_file(
+        repo_id=repo_id,
+        filename="tokens.txt",
+        subfolder=".",
+    )
+
+    tts_config = sherpa_onnx.OfflineTtsConfig(
+        model=sherpa_onnx.OfflineTtsModelConfig(
+            vits=sherpa_onnx.OfflineTtsVitsModelConfig(
+                model=model,
+                lexicon=lexicon,
+                tokens=tokens,
+                length_scale=1.0 / speed,
+            ),
+            matcha=sherpa_onnx.OfflineTtsMatchaModelConfig(),
+            provider="cpu",
+            debug=True,
+            num_threads=2,
+        ),
+        max_num_sentences=1,
+    )
+    tts = sherpa_onnx.OfflineTts(tts_config)
+
+    return tts
+
+
+@lru_cache(maxsize=10)
+def _get_vits_ljs(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    assert repo_id == "csukuangfj/vits-ljs"
+
+    model = get_file(
+        repo_id=repo_id,
+        filename="vits-ljs.onnx",
+        subfolder=".",
+    )
+
+    lexicon = get_file(
+        repo_id=repo_id,
+        filename="lexicon.txt",
+        subfolder=".",
+    )
+
+    tokens = get_file(
+        repo_id=repo_id,
+        filename="tokens.txt",
+        subfolder=".",
+    )
+
+    tts_config = sherpa_onnx.OfflineTtsConfig(
+        model=sherpa_onnx.OfflineTtsModelConfig(
+            vits=sherpa_onnx.OfflineTtsVitsModelConfig(
+                model=model,
+                lexicon=lexicon,
+                tokens=tokens,
+                length_scale=1.0 / speed,
+            ),
+            matcha=sherpa_onnx.OfflineTtsMatchaModelConfig(),
+            provider="cpu",
+            debug=True,
+            num_threads=2,
+        ),
+        max_num_sentences=1,
+    )
+    tts = sherpa_onnx.OfflineTts(tts_config)
+
+    return tts
+
+
+@lru_cache(maxsize=10)
+def _get_kokoro(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    data_dir = "/tmp/espeak-ng-data"
+    repo_id = repo_id.split("|")[0]
+    assert repo_id in (
+        "csukuangfj/kokoro-en-v0_19",
+        "csukuangfj/kokoro-multi-lang-v1_0",
+        "csukuangfj/kokoro-multi-lang-v1_1",
+    ), repo_id
+
+    model = get_file(
+        repo_id=repo_id,
+        filename="model.onnx",
+        subfolder=".",
+    )
+
+    tokens = get_file(
+        repo_id=repo_id,
+        filename="tokens.txt",
+        subfolder=".",
+    )
+
+    voices = get_file(
+        repo_id=repo_id,
+        filename="voices.bin",
+        subfolder=".",
+    )
+    if repo_id in (
+        "csukuangfj/kokoro-multi-lang-v1_0",
+        "csukuangfj/kokoro-multi-lang-v1_1",
+    ):
+        lexicon_en = get_file(
+            repo_id=repo_id,
+            filename="lexicon-us-en.txt",
+            subfolder=".",
+        )
+        lexicon_zh = get_file(
+            repo_id=repo_id,
+            filename="lexicon-zh.txt",
+            subfolder=".",
+        )
+        lexicon = f"{lexicon_en},{lexicon_zh}"
+
+        date_zh = get_file(
+            repo_id=repo_id,
+            filename="date-zh.fst",
+            subfolder=".",
+        )
+
+        number_zh = get_file(
+            repo_id=repo_id,
+            filename="number-zh.fst",
+            subfolder=".",
+        )
+        phone_zh = get_file(
+            repo_id=repo_id,
+            filename="phone-zh.fst",
+            subfolder=".",
+        )
+        rule_fsts = f"{date_zh},{phone_zh},{number_zh}"
+        dict_dir = "/tmp/dict"
+    else:
+        lexicon = ""
+        rule_fsts = ""
+        dict_dir = ""
+
+    tts_config = sherpa_onnx.OfflineTtsConfig(
+        model=sherpa_onnx.OfflineTtsModelConfig(
+            kokoro=sherpa_onnx.OfflineTtsKokoroModelConfig(
+                model=model,
+                voices=voices,
+                tokens=tokens,
+                data_dir=data_dir,
+                length_scale=1.0 / speed,
+                lexicon=lexicon,
+                dict_dir=dict_dir,
+            ),
+            provider="cpu",
+            debug=True,
+            num_threads=2,
+        ),
+        max_num_sentences=1,
+        rule_fsts=rule_fsts,
+    )
+
+    tts = sherpa_onnx.OfflineTts(tts_config)
+
+    return tts
+
+
+@lru_cache(maxsize=10)
+def _get_vits_piper(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    data_dir = "/tmp/espeak-ng-data"
+    repo_id = repo_id.split("|")[0]
+
+    if "coqui" in repo_id or "vits-mms" in repo_id:
+        name = "model"
+    elif "piper" in repo_id:
+        n = len("vits-piper-")
+        name = repo_id.split("/")[1][n:]
+    elif "mimic3" in repo_id:
+        n = len("vits-mimic3-")
+        name = repo_id.split("/")[1][n:]
+    else:
+        raise ValueError(f"Unsupported {repo_id}")
+
+    if "vits-coqui-uk-mai" in repo_id or "vits-mms" in repo_id:
+        data_dir = ""
+
+    model = get_file(
+        repo_id=repo_id,
+        filename=f"{name}.onnx",
+        subfolder=".",
+    )
+
+    tokens = get_file(
+        repo_id=repo_id,
+        filename="tokens.txt",
+        subfolder=".",
+    )
+
+    tts_config = sherpa_onnx.OfflineTtsConfig(
+        model=sherpa_onnx.OfflineTtsModelConfig(
+            vits=sherpa_onnx.OfflineTtsVitsModelConfig(
+                model=model,
+                lexicon="",
+                data_dir=data_dir,
+                tokens=tokens,
+                length_scale=1.0 / speed,
+            ),
+            matcha=sherpa_onnx.OfflineTtsMatchaModelConfig(),
+            provider="cpu",
+            debug=True,
+            num_threads=2,
+        ),
+        max_num_sentences=1,
+    )
+    tts = sherpa_onnx.OfflineTts(tts_config)
+
+    return tts
+
+
+@lru_cache(maxsize=10)
+def _get_vits_mms(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    return _get_vits_piper(repo_id, speed)
+
+
+@lru_cache(maxsize=10)
+def _get_vits_zh_aishell3(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    repo_id = repo_id.split("|")[0]
+    assert repo_id == "csukuangfj/vits-zh-aishell3", repo_id
+
+    model = get_file(
+        repo_id=repo_id,
+        filename="vits-aishell3.onnx",
+        subfolder=".",
+    )
+
+    lexicon = get_file(
+        repo_id=repo_id,
+        filename="lexicon.txt",
+        subfolder=".",
+    )
+
+    tokens = get_file(
+        repo_id=repo_id,
+        filename="tokens.txt",
+        subfolder=".",
+    )
+
+    rule_fsts = ["phone.fst", "date.fst", "number.fst", "new_heteronym.fst"]
+
+    rule_fsts = [
+        get_file(
+            repo_id=repo_id,
+            filename=f,
+            subfolder=".",
+        )
+        for f in rule_fsts
+    ]
+    rule_fsts = ",".join(rule_fsts)
+
+    rule_fars = get_file(
+        repo_id=repo_id,
+        filename="rule.far",
+        subfolder=".",
+    )
+
+    tts_config = sherpa_onnx.OfflineTtsConfig(
+        model=sherpa_onnx.OfflineTtsModelConfig(
+            vits=sherpa_onnx.OfflineTtsVitsModelConfig(
+                model=model,
+                lexicon=lexicon,
+                tokens=tokens,
+                length_scale=1.0 / speed,
+            ),
+            matcha=sherpa_onnx.OfflineTtsMatchaModelConfig(),
+            provider="cpu",
+            debug=True,
+            num_threads=2,
+        ),
+        rule_fsts=rule_fsts,
+        rule_fars=rule_fars,
+        max_num_sentences=1,
+    )
+    tts = sherpa_onnx.OfflineTts(tts_config)
+
+    return tts
+
+
+@lru_cache(maxsize=10)
+def _get_matcha_hf_espeak(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    repo_id = repo_id.split("|")[0]
+    assert repo_id in (
+        "csukuangfj/matcha-tts-fa_en-khadijah",
+        "csukuangfj/matcha-tts-fa_en-musa",
+    ), repo_id
+
+    acoustic_model = get_file(
+        repo_id=repo_id,
+        filename="model.onnx",
+        subfolder=".",
+    )
+
+    vocoder = get_file(
+        repo_id="csukuangfj/sherpa-onnx-hifigan",
+        filename="hifigan_v2.onnx",
+        subfolder=".",
+    )
+
+    tokens = get_file(
+        repo_id=repo_id,
+        filename="tokens.txt",
+        subfolder=".",
+    )
+
+    data_dir = "/tmp/espeak-ng-data"
+    tts_config = sherpa_onnx.OfflineTtsConfig(
+        model=sherpa_onnx.OfflineTtsModelConfig(
+            vits=sherpa_onnx.OfflineTtsVitsModelConfig(),
+            matcha=sherpa_onnx.OfflineTtsMatchaModelConfig(
+                acoustic_model=acoustic_model,
+                vocoder=vocoder,
+                tokens=tokens,
+                lexicon="",
+                data_dir=data_dir,
+                length_scale=1.0 / speed,
+            ),
+            provider="cpu",
+            debug=True,
+            num_threads=2,
+        ),
+        max_num_sentences=1,
+    )
+    tts = sherpa_onnx.OfflineTts(tts_config)
+
+    return tts
+
+
+@lru_cache(maxsize=10)
+def _get_matcha_hf(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    repo_id = repo_id.split("|")[0]
+    assert repo_id in ("csukuangfj/matcha-icefall-zh-baker",), repo_id
+
+    if repo_id == "csukuangfj/matcha-icefall-zh-baker":
+        acoustic_model = "model-steps-3.onnx"
+
+    if not Path("/tmp/dict").is_dir():
+        os.system(
+            "cd /tmp; curl -SL -O https://github.com/csukuangfj/cppjieba/releases/download/sherpa-onnx-2024-04-19/dict.tar.bz2; tar xvf dict.tar.bz2"
+        )
+    os.system("ls -lh /tmp/dict")
+
+    acoustic_model = get_file(
+        repo_id=repo_id,
+        filename=acoustic_model,
+        subfolder=".",
+    )
+
+    vocoder = get_file(
+        repo_id="csukuangfj/sherpa-onnx-hifigan",
+        filename="hifigan_v2.onnx",
+        subfolder=".",
+    )
+
+    lexicon = get_file(
+        repo_id=repo_id,
+        filename="lexicon.txt",
+        subfolder=".",
+    )
+
+    tokens = get_file(
+        repo_id=repo_id,
+        filename="tokens.txt",
+        subfolder=".",
+    )
+
+    rule_fars = ""
+
+    rule_fsts = ["phone.fst", "date.fst", "number.fst"]
+
+    rule_fsts = [
+        get_file(
+            repo_id=repo_id,
+            filename=f,
+            subfolder=".",
+        )
+        for f in rule_fsts
+    ]
+    rule_fsts = ",".join(rule_fsts)
+
+    dict_dir = "/tmp/dict"
+
+    tts_config = sherpa_onnx.OfflineTtsConfig(
+        model=sherpa_onnx.OfflineTtsModelConfig(
+            vits=sherpa_onnx.OfflineTtsVitsModelConfig(),
+            matcha=sherpa_onnx.OfflineTtsMatchaModelConfig(
+                acoustic_model=acoustic_model,
+                vocoder=vocoder,
+                lexicon=lexicon,
+                tokens=tokens,
+                dict_dir=dict_dir,
+                length_scale=1.0 / speed,
+            ),
+            provider="cpu",
+            debug=True,
+            num_threads=2,
+        ),
+        rule_fsts=rule_fsts,
+        rule_fars=rule_fars,
+        max_num_sentences=1,
+    )
+    tts = sherpa_onnx.OfflineTts(tts_config)
+
+    return tts
+
+
+@lru_cache(maxsize=10)
+def _get_vits_hf(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    repo_id = repo_id.split("|")[0]
+
+    if "fanchen" in repo_id or "vits-cantonese-hf-xiaomaiiwn" in repo_id:
+        model = repo_id.split("/")[-1]
+    elif "csukuangfj/vits-melo-tts-zh_en" == repo_id:
+        model = "model"
+    else:
+        model = repo_id.split("-")[-1]
+
+    if "sherpa-onnx-vits-zh-ll" in repo_id:
+        model = "model"
+
+    if not Path("/tmp/dict").is_dir():
+        os.system(
+            "cd /tmp; curl -SL -O https://github.com/csukuangfj/cppjieba/releases/download/sherpa-onnx-2024-04-19/dict.tar.bz2; tar xvf dict.tar.bz2"
+        )
+    os.system("ls -lh /tmp/dict")
+
+    model = get_file(
+        repo_id=repo_id,
+        filename=f"{model}.onnx",
+        subfolder=".",
+    )
+
+    lexicon = get_file(
+        repo_id=repo_id,
+        filename="lexicon.txt",
+        subfolder=".",
+    )
+
+    tokens = get_file(
+        repo_id=repo_id,
+        filename="tokens.txt",
+        subfolder=".",
+    )
+
+    rule_fars = ""
+
+    if "vits-cantonese-hf-xiaomaiiwn" not in repo_id:
+        rule_fsts = ["phone.fst", "date.fst", "number.fst"]
+
+        rule_fsts = [
+            get_file(
+                repo_id=repo_id,
+                filename=f,
+                subfolder=".",
+            )
+            for f in rule_fsts
+        ]
+        rule_fsts = ",".join(rule_fsts)
+
+        #  rule_fars = get_file(
+        #      repo_id=repo_id,
+        #      filename="rule.far",
+        #      subfolder=".",
+        #  )
+        vits_dict_dir = "/tmp/dict"
+    else:
+        rule_fsts = get_file(
+            repo_id=repo_id,
+            filename="rule.fst",
+            subfolder=".",
+        )
+        vits_dict_dir = ""
+
+    tts_config = sherpa_onnx.OfflineTtsConfig(
+        model=sherpa_onnx.OfflineTtsModelConfig(
+            vits=sherpa_onnx.OfflineTtsVitsModelConfig(
+                model=model,
+                lexicon=lexicon,
+                tokens=tokens,
+                dict_dir=vits_dict_dir,
+                length_scale=1.0 / speed,
+            ),
+            matcha=sherpa_onnx.OfflineTtsMatchaModelConfig(),
+            provider="cpu",
+            debug=True,
+            num_threads=2,
+        ),
+        rule_fsts=rule_fsts,
+        rule_fars=rule_fars,
+        max_num_sentences=1,
+    )
+    tts = sherpa_onnx.OfflineTts(tts_config)
+
+    return tts
+
+
+@lru_cache(maxsize=10)
+def get_pretrained_model(repo_id: str, speed: float) -> sherpa_onnx.OfflineTts:
+    if repo_id in chinese_models:
+        return chinese_models[repo_id](repo_id, speed)
+    elif repo_id in chinese_english_models:
+        return chinese_english_models[repo_id](repo_id, speed)
+    elif repo_id in persian_english_models:
+        return persian_english_models[repo_id](repo_id, speed)
+    if repo_id in cantonese_models:
+        return cantonese_models[repo_id](repo_id, speed)
+    elif repo_id in english_models:
+        return english_models[repo_id](repo_id, speed)
+    elif repo_id in german_models:
+        return german_models[repo_id](repo_id, speed)
+    elif repo_id in spanish_models:
+        return spanish_models[repo_id](repo_id, speed)
+    elif repo_id in french_models:
+        return french_models[repo_id](repo_id, speed)
+    elif repo_id in ukrainian_models:
+        return ukrainian_models[repo_id](repo_id, speed)
+    elif repo_id in russian_models:
+        return russian_models[repo_id](repo_id, speed)
+    elif repo_id in arabic_models:
+        return arabic_models[repo_id](repo_id, speed)
+    elif repo_id in catalan_models:
+        return catalan_models[repo_id](repo_id, speed)
+    elif repo_id in czech_models:
+        return czech_models[repo_id](repo_id, speed)
+    elif repo_id in danish_models:
+        return danish_models[repo_id](repo_id, speed)
+    elif repo_id in greek_models:
+        return greek_models[repo_id](repo_id, speed)
+    elif repo_id in finnish_models:
+        return finnish_models[repo_id](repo_id, speed)
+    elif repo_id in hungarian_models:
+        return hungarian_models[repo_id](repo_id, speed)
+    elif repo_id in icelandic_models:
+        return icelandic_models[repo_id](repo_id, speed)
+    elif repo_id in italian_models:
+        return italian_models[repo_id](repo_id, speed)
+    elif repo_id in georgian_models:
+        return georgian_models[repo_id](repo_id, speed)
+    elif repo_id in kazakh_models:
+        return kazakh_models[repo_id](repo_id, speed)
+    elif repo_id in luxembourgish_models:
+        return luxembourgish_models[repo_id](repo_id, speed)
+    elif repo_id in nepali_models:
+        return nepali_models[repo_id](repo_id, speed)
+    elif repo_id in dutch_models:
+        return dutch_models[repo_id](repo_id, speed)
+    elif repo_id in norwegian_models:
+        return norwegian_models[repo_id](repo_id, speed)
+    elif repo_id in polish_models:
+        return polish_models[repo_id](repo_id, speed)
+    elif repo_id in portuguese_models:
+        return portuguese_models[repo_id](repo_id, speed)
+    elif repo_id in romanian_models:
+        return romanian_models[repo_id](repo_id, speed)
+    elif repo_id in slovak_models:
+        return slovak_models[repo_id](repo_id, speed)
+    elif repo_id in serbian_models:
+        return serbian_models[repo_id](repo_id, speed)
+    elif repo_id in swedish_models:
+        return swedish_models[repo_id](repo_id, speed)
+    elif repo_id in swahili_models:
+        return swahili_models[repo_id](repo_id, speed)
+    elif repo_id in turkish_models:
+        return turkish_models[repo_id](repo_id, speed)
+    elif repo_id in vietnamese_models:
+        return vietnamese_models[repo_id](repo_id, speed)
+    elif repo_id in bulgarian_models:
+        return bulgarian_models[repo_id](repo_id, speed)
+    elif repo_id in estonian_models:
+        return estonian_models[repo_id](repo_id, speed)
+    elif repo_id in irish_models:
+        return irish_models[repo_id](repo_id, speed)
+    elif repo_id in croatian_models:
+        return croatian_models[repo_id](repo_id, speed)
+    elif repo_id in lithuanian_models:
+        return lithuanian_models[repo_id](repo_id, speed)
+    elif repo_id in latvian_models:
+        return latvian_models[repo_id](repo_id, speed)
+    elif repo_id in maltese_models:
+        return maltese_models[repo_id](repo_id, speed)
+    elif repo_id in slovenian_models:
+        return slovenian_models[repo_id](repo_id, speed)
+    elif repo_id in bengali_models:
+        return bengali_models[repo_id](repo_id, speed)
+    elif repo_id in min_nan_models:
+        return min_nan_models[repo_id](repo_id, speed)
+    elif repo_id in thai_models:
+        return thai_models[repo_id](repo_id, speed)
+    elif repo_id in persian_models:
+        return persian_models[repo_id](repo_id, speed)
+    elif repo_id in korean_models:
+        return korean_models[repo_id](repo_id, speed)
+    elif repo_id in afrikaans_models:
+        return afrikaans_models[repo_id](repo_id, speed)
+    elif repo_id in gujarati_models:
+        return gujarati_models[repo_id](repo_id, speed)
+    elif repo_id in tswana_models:
+        return tswana_models[repo_id](repo_id, speed)
+    elif repo_id in welsh_models:
+        return welsh_models[repo_id](repo_id, speed)
+    else:
+        raise ValueError(f"Unsupported repo_id: {repo_id}")
+
+
+cantonese_models = {
+    "csukuangfj/vits-cantonese-hf-xiaomaiiwn": _get_vits_hf,
+}
+
+chinese_english_models = {
+    "csukuangfj/kokoro-multi-lang-v1_1|103 speakers": _get_kokoro,
+    "csukuangfj/kokoro-multi-lang-v1_0|53 speakers": _get_kokoro,
+    "csukuangfj/vits-melo-tts-zh_en|1": _get_vits_hf,  # 1
+}
+
+persian_english_models = {
+    "csukuangfj/matcha-tts-fa_en-khadijah|1 speaker": _get_matcha_hf_espeak,  # 1
+    "csukuangfj/matcha-tts-fa_en-musa|1 speaker": _get_matcha_hf_espeak,  # 1
+    "csukuangfj/vits-piper-fa_en-rezahedayatfar-ibrahimwalk-medium|1": _get_vits_piper,  # 1
+}
+
+chinese_models = {
+    "csukuangfj/matcha-icefall-zh-baker|1 speaker": _get_matcha_hf,  # 1
+    "csukuangfj/vits-zh-hf-fanchen-wnj|1 speaker": _get_vits_hf,  # 1
+    "csukuangfj/vits-zh-hf-fanchen-C|187 speakers": _get_vits_hf,  # 187
+    "csukuangfj/sherpa-onnx-vits-zh-ll|5 speakers": _get_vits_hf,  # 804
+    "csukuangfj/vits-zh-hf-keqing|804 speakers": _get_vits_hf,  # 804
+    "csukuangfj/vits-zh-hf-theresa|804 speakers": _get_vits_hf,  # 804
+    "csukuangfj/vits-zh-hf-eula|804 speakers": _get_vits_hf,  # 804
+    "csukuangfj/vits-zh-hf-echo|804 speakers": _get_vits_hf,  # 804
+    "csukuangfj/vits-zh-hf-bronya|804 speakers": _get_vits_hf,  # 804
+    "csukuangfj/vits-zh-hf-doom|804 speakers": _get_vits_hf,  # 804
+    "csukuangfj/vits-zh-hf-zenyatta|804 speakers": _get_vits_hf,  # 804
+    "csukuangfj/vits-zh-hf-abyssinvoker|804 speakers": _get_vits_hf,  # 804
+    "csukuangfj/vits-zh-hf-fanchen-ZhiHuiLaoZhe|1 speaker": _get_vits_hf,  # 1
+    "csukuangfj/vits-zh-hf-fanchen-ZhiHuiLaoZhe_new|1 speaker": _get_vits_hf,  # 1
+    "csukuangfj/vits-zh-hf-fanchen-unity|1 speaker": _get_vits_hf,  # 1
+    "csukuangfj/vits-zh-aishell3|174 speakers": _get_vits_zh_aishell3,
+    "csukuangfj/vits-piper-zh_CN-huayan-medium|1 speaker": _get_vits_piper,
+    #  "csukuangfj/vits-piper-zh_CN-huayan-x_low": _get_vits_piper,
+}
+
+english_models = {
+    "csukuangfj/kokoro-en-v0_19|11 speakers": _get_kokoro,
+    "csukuangfj/vits-piper-en_US-glados-high|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-glados|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-southern_english_male-medium|8 speakers": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-southern_english_female-medium|6 speakers": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-bryce-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-john-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-norman-medium|1 speaker": _get_vits_piper,
+    # coqui-ai
+    "csukuangfj/vits-coqui-en-ljspeech|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-coqui-en-ljspeech-neon|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-coqui-en-vctk|109 speakers": _get_vits_piper,
+    # piper, US
+    "csukuangfj/vits-piper-en_GB-sweetbbak-amy|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-amy-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-amy-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-arctic-medium|18 speakers": _get_vits_piper,  # 18 speakers
+    "csukuangfj/vits-piper-en_US-danny-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-hfc_male-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-hfc_female-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-joe-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-kathleen-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-kusal-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-l2arctic-medium|24 speakers": _get_vits_piper,  # 24 speakers
+    "csukuangfj/vits-piper-en_US-lessac-high|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-lessac-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-lessac-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-libritts-high|904 speakers": _get_vits_piper,  # 904 speakers
+    "csukuangfj/vits-piper-en_US-libritts_r-medium|904 speakers": _get_vits_piper,  # 904 speakers
+    "csukuangfj/vits-piper-en_US-ljspeech-high|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-ljspeech-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-ryan-high|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-ryan-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_US-ryan-medium|1 speaker": _get_vits_piper,
+    # piper, GB
+    "csukuangfj/vits-piper-en_GB-alan-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-alan-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-alan-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-cori-high|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-cori-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-jenny_dioco-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-northern_english_male-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-semaine-medium|4 speakers": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-southern_english_female-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-en_GB-vctk-medium|109 speakers": _get_vits_piper,
+    #
+    "csukuangfj/vits-vctk|109 speakers": _get_vits_vctk,  # 109 speakers
+    "csukuangfj/vits-ljs|1 speaker": _get_vits_ljs,
+}
+
+german_models = {
+    "csukuangfj/vits-piper-de_DE-glados-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-glados-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-glados-high|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-coqui-de-css10|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-eva_k-x_low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-karlsson-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-kerstin-low|1 speaker": _get_vits_piper,
+    #  "csukuangfj/vits-piper-de_DE-mls-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-pavoque-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-ramona-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-thorsten-low|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-thorsten-medium|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-thorsten-high|1 speaker": _get_vits_piper,
+    "csukuangfj/vits-piper-de_DE-thorsten_emotional-medium|8 speakers": _get_vits_piper,  # 8 speakers
+}
+
+spanish_models = {
+    #  "csukuangfj/vits-coqui-es-css10": _get_vits_piper,
+    "csukuangfj/vits-piper-es-glados-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-es_ES-carlfm-x_low": _get_vits_piper,
+    "csukuangfj/vits-piper-es_ES-davefx-medium": _get_vits_piper,
+    #  "csukuangfj/vits-piper-es_ES-mls_10246-low": _get_vits_piper,
+    #  "csukuangfj/vits-piper-es_ES-mls_9972-low": _get_vits_piper,
+    "csukuangfj/vits-piper-es_ES-sharvard-medium": _get_vits_piper,  # 2 speakers
+    "csukuangfj/vits-piper-es_MX-ald-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-es_MX-claude-high": _get_vits_piper,
+    "csukuangfj/vits-mimic3-es_ES-m-ailabs_low": _get_vits_piper,
+}
+
+french_models = {
+    "csukuangfj/vits-coqui-fr-css10": _get_vits_piper,
+    #  "csukuangfj/vits-piper-fr_FR-gilles-low": _get_vits_piper,
+    #  "csukuangfj/vits-piper-fr_FR-mls_1840-low": _get_vits_piper,
+    #  "csukuangfj/vits-piper-fr_FR-mls-medium": _get_vits_piper,  # 2 speakers, 0-femal, 1-male
+    "csukuangfj/vits-piper-fr_FR-upmc-medium": _get_vits_piper,  # 2 speakers, 0-femal, 1-male
+    "csukuangfj/vits-piper-fr_FR-tom-medium|1 speaker": _get_vits_piper,  # 2 speakers, 0-femal, 1-male
+    "csukuangfj/vits-piper-fr_FR-siwis-low": _get_vits_piper,  # female
+    "csukuangfj/vits-piper-fr_FR-siwis-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-fr_FR-tjiho-model1": _get_vits_piper,
+    "csukuangfj/vits-piper-fr_FR-tjiho-model2": _get_vits_piper,
+    "csukuangfj/vits-piper-fr_FR-tjiho-model3": _get_vits_piper,
+}
+
+ukrainian_models = {
+    "csukuangfj/vits-piper-uk_UA-lada-x_low": _get_vits_piper,
+    "csukuangfj/vits-coqui-uk-mai": _get_vits_piper,
+    #  "csukuangfj/vits-piper-uk_UA-ukrainian_tts-medium": _get_vits_piper, # does not work somehow
+}
+
+russian_models = {
+    "csukuangfj/vits-piper-ru_RU-denis-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-ru_RU-dmitri-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-ru_RU-irina-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-ru_RU-ruslan-medium": _get_vits_piper,
+}
+
+arabic_models = {
+    "csukuangfj/vits-piper-ar_JO-kareem-low": _get_vits_piper,
+    "csukuangfj/vits-piper-ar_JO-kareem-medium": _get_vits_piper,
+}
+
+catalan_models = {
+    "csukuangfj/vits-piper-ca_ES-upc_ona-x_low": _get_vits_piper,
+    "csukuangfj/vits-piper-ca_ES-upc_ona-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-ca_ES-upc_pau-x_low": _get_vits_piper,
+}
+
+czech_models = {
+    "csukuangfj/vits-piper-cs_CZ-jirka-low": _get_vits_piper,
+    "csukuangfj/vits-piper-cs_CZ-jirka-medium": _get_vits_piper,
+    "csukuangfj/vits-coqui-cs-cv": _get_vits_piper,
+}
+
+danish_models = {
+    "csukuangfj/vits-coqui-da-cv": _get_vits_piper,
+    "csukuangfj/vits-piper-da_DK-talesyntese-medium": _get_vits_piper,
+}
+
+greek_models = {
+    "csukuangfj/vits-piper-el_GR-rapunzelina-low": _get_vits_piper,
+    #  "csukuangfj/vits-mimic3-el_GR-rapunzelina_low": _get_vits_piper,
+}
+
+finnish_models = {
+    "csukuangfj/vits-coqui-fi-css10": _get_vits_piper,
+    "csukuangfj/vits-piper-fi_FI-harri-low": _get_vits_piper,
+    "csukuangfj/vits-piper-fi_FI-harri-medium": _get_vits_piper,
+    "csukuangfj/vits-mimic3-fi_FI-harri-tapani-ylilammi_low": _get_vits_piper,
+}
+
+hungarian_models = {
+    #  "csukuangfj/vits-coqui-hu-css10": _get_vits_piper,
+    "csukuangfj/vits-piper-hu_HU-anna-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-hu_HU-berta-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-hu_HU-imre-medium": _get_vits_piper,
+    "csukuangfj/vits-mimic3-hu_HU-diana-majlinger_low": _get_vits_piper,
+}
+
+icelandic_models = {
+    "csukuangfj/vits-piper-is_IS-bui-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-is_IS-salka-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-is_IS-steinn-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-is_IS-ugla-medium": _get_vits_piper,
+}
+
+italian_models = {
+    "csukuangfj/vits-piper-it_IT-riccardo-x_low": _get_vits_piper,
+    "csukuangfj/vits-piper-it_IT-paola-medium": _get_vits_piper,
+}
+
+georgian_models = {
+    "csukuangfj/vits-piper-ka_GE-natia-medium": _get_vits_piper,
+}
+
+kazakh_models = {
+    "csukuangfj/vits-piper-kk_KZ-iseke-x_low": _get_vits_piper,
+    "csukuangfj/vits-piper-kk_KZ-issai-high": _get_vits_piper,
+    "csukuangfj/vits-piper-kk_KZ-raya-x_low": _get_vits_piper,
+}
+
+luxembourgish_models = {
+    "csukuangfj/vits-piper-lb_LU-marylux-medium": _get_vits_piper,
+}
+
+nepali_models = {
+    "csukuangfj/vits-piper-ne_NP-google-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-ne_NP-google-x_low": _get_vits_piper,
+    "csukuangfj/vits-mimic3-ne_NP-ne-google_low": _get_vits_piper,
+}
+
+dutch_models = {
+    "csukuangfj/vits-coqui-nl-css10": _get_vits_piper,
+    "csukuangfj/vits-piper-nl_BE-nathalie-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-nl_BE-nathalie-x_low": _get_vits_piper,
+    "csukuangfj/vits-piper-nl_BE-rdh-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-nl_BE-rdh-x_low": _get_vits_piper,
+    #  "csukuangfj/vits-piper-nl_NL-mls-medium": _get_vits_piper,
+    #  "csukuangfj/vits-piper-nl_NL-mls_5809-low": _get_vits_piper,
+    #  "csukuangfj/vits-piper-nl_NL-mls_7432-low": _get_vits_piper,
+}
+
+norwegian_models = {
+    "csukuangfj/vits-piper-no_NO-talesyntese-medium": _get_vits_piper,
+}
+
+polish_models = {
+    "csukuangfj/vits-coqui-pl-mai_female": _get_vits_piper,
+    "csukuangfj/vits-piper-pl_PL-darkman-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-pl_PL-gosia-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-pl_PL-mc_speech-medium": _get_vits_piper,
+    #  "csukuangfj/vits-piper-pl_PL-mls_6892-low": _get_vits_piper,
+    "csukuangfj/vits-mimic3-pl_PL-m-ailabs_low": _get_vits_piper,
+}
+
+portuguese_models = {
+    "csukuangfj/vits-coqui-pt-cv": _get_vits_piper,
+    "csukuangfj/vits-piper-pt_BR-edresson-low": _get_vits_piper,
+    "csukuangfj/vits-piper-pt_BR-faber-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-pt_PT-tugao-medium": _get_vits_piper,
+}
+
+romanian_models = {
+    "csukuangfj/vits-coqui-ro-cv": _get_vits_piper,
+    "csukuangfj/vits-piper-ro_RO-mihai-medium": _get_vits_piper,
+}
+
+
+slovak_models = {
+    "csukuangfj/vits-coqui-sk-cv": _get_vits_piper,
+    "csukuangfj/vits-piper-sk_SK-lili-medium": _get_vits_piper,
+}
+
+serbian_models = {
+    "csukuangfj/vits-piper-sr_RS-serbski_institut-medium": _get_vits_piper,
+}
+
+swedish_models = {
+    "csukuangfj/vits-coqui-sv-cv": _get_vits_piper,
+    "csukuangfj/vits-piper-sv_SE-nst-medium": _get_vits_piper,
+}
+
+swahili_models = {
+    "csukuangfj/vits-piper-sw_CD-lanfrica-medium": _get_vits_piper,
+}
+
+turkish_models = {
+    "csukuangfj/vits-piper-tr_TR-dfki-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-tr_TR-fahrettin-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-tr_TR-fettah-medium|1 speaker": _get_vits_piper,
+}
+
+vietnamese_models = {
+    "csukuangfj/vits-piper-vi_VN-25hours_single-low": _get_vits_piper,
+    "csukuangfj/vits-piper-vi_VN-vais1000-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-vi_VN-vivos-x_low": _get_vits_piper,
+    "csukuangfj/vits-mimic3-vi_VN-vais1000_low": _get_vits_piper,
+}
+
+bulgarian_models = {
+    "csukuangfj/vits-coqui-bg-cv": _get_vits_piper,
+}
+
+estonian_models = {
+    "csukuangfj/vits-coqui-et-cv": _get_vits_piper,
+}
+
+irish_models = {
+    "csukuangfj/vits-coqui-ga-cv": _get_vits_piper,
+}
+
+croatian_models = {
+    "csukuangfj/vits-coqui-hr-cv": _get_vits_piper,
+}
+
+lithuanian_models = {
+    "csukuangfj/vits-coqui-lt-cv": _get_vits_piper,
+}
+
+latvian_models = {
+    "csukuangfj/vits-piper-lv_LV-aivars-medium": _get_vits_piper,
+    "csukuangfj/vits-coqui-lv-cv": _get_vits_piper,
+}
+
+maltese_models = {
+    "csukuangfj/vits-coqui-mt-cv": _get_vits_piper,
+}
+
+slovenian_models = {
+    "csukuangfj/vits-piper-sl_SI-artur-medium": _get_vits_piper,
+    "csukuangfj/vits-coqui-sl-cv": _get_vits_piper,
+}
+
+# Bangla
+bengali_models = {
+    "csukuangfj/vits-coqui-bn-custom_female": _get_vits_piper,
+    "csukuangfj/vits-mimic3-bn-multi_low": _get_vits_piper,
+}
+
+min_nan_models = {
+    "csukuangfj/vits-mms-nan": _get_vits_mms,
+}
+
+thai_models = {
+    "csukuangfj/vits-mms-tha": _get_vits_mms,
+}
+
+persian_models = {
+    "csukuangfj/vits-piper-fa_IR-amir-medium": _get_vits_piper,
+    "csukuangfj/vits-piper-fa_IR-gyro-medium": _get_vits_piper,
+    "csukuangfj/vits-mimic3-fa-haaniye_low": _get_vits_piper,
+}
+
+korean_models = {
+    "csukuangfj/vits-mimic3-ko_KO-kss_low": _get_vits_piper,
+}
+
+
+afrikaans_models = {
+    "csukuangfj/vits-mimic3-af_ZA-google-nwu_low": _get_vits_piper,
+}
+
+gujarati_models = {
+    "csukuangfj/vits-mimic3-gu_IN-cmu-indic_low": _get_vits_piper,
+}
+
+tswana_models = {
+    "csukuangfj/vits-mimic3-tn_ZA-google-nwu_low": _get_vits_piper,
+}
+
+welsh_models = {
+    "csukuangfj/vits-piper-cy_GB-gwryw_gogleddol-medium|1 speaker": _get_vits_piper,
+}
+
+language_to_models = {
+    "English": list(english_models.keys()),
+    "Chinese (Mandarin, 普通话)": list(chinese_models.keys()),
+    "Chinese+English": list(chinese_english_models.keys()),
+    "Persian+English": list(persian_english_models.keys()),
+    "Cantonese (粤语)": list(cantonese_models.keys()),
+    "Min-nan (闽南话)": list(min_nan_models.keys()),
+    "Arabic": list(arabic_models.keys()),
+    "Afrikaans": list(afrikaans_models.keys()),
+    "Bengali": list(bengali_models.keys()),
+    "Bulgarian": list(bulgarian_models.keys()),
+    "Catalan": list(catalan_models.keys()),
+    "Croatian": list(croatian_models.keys()),
+    "Czech": list(czech_models.keys()),
+    "Danish": list(danish_models.keys()),
+    "Dutch": list(dutch_models.keys()),
+    "Estonian": list(estonian_models.keys()),
+    "Finnish": list(finnish_models.keys()),
+    "French": list(french_models.keys()),
+    "Georgian": list(georgian_models.keys()),
+    "German": list(german_models.keys()),
+    "Greek": list(greek_models.keys()),
+    "Gujarati": list(gujarati_models.keys()),
+    "Hungarian": list(hungarian_models.keys()),
+    "Icelandic": list(icelandic_models.keys()),
+    "Irish": list(irish_models.keys()),
+    "Italian": list(italian_models.keys()),
+    "Kazakh": list(kazakh_models.keys()),
+    "Korean": list(korean_models.keys()),
+    "Latvian": list(latvian_models.keys()),
+    "Lithuanian": list(lithuanian_models.keys()),
+    "Luxembourgish": list(luxembourgish_models.keys()),
+    "Maltese": list(maltese_models.keys()),
+    "Nepali": list(nepali_models.keys()),
+    "Norwegian": list(norwegian_models.keys()),
+    "Persian": list(persian_models.keys()),
+    "Polish": list(polish_models.keys()),
+    "Portuguese": list(portuguese_models.keys()),
+    "Romanian": list(romanian_models.keys()),
+    "Russian": list(russian_models.keys()),
+    "Serbian": list(serbian_models.keys()),
+    "Slovak": list(slovak_models.keys()),
+    "Slovenian": list(slovenian_models.keys()),
+    "Spanish": list(spanish_models.keys()),
+    "Swahili": list(swahili_models.keys()),
+    "Swedish": list(swedish_models.keys()),
+    "Thai": list(thai_models.keys()),
+    "Tswana": list(tswana_models.keys()),
+    "Turkish": list(turkish_models.keys()),
+    "Ukrainian": list(ukrainian_models.keys()),
+    "Vietnamese": list(vietnamese_models.keys()),
+    "Welsh": list(welsh_models.keys()),
+}

requirements.txt

@@ -0,0 +1,4 @@
+https://huggingface.co/csukuangfj/sherpa-onnx-wheels/resolve/main/cpu/1.10.46/sherpa_onnx-1.10.46-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
+#sherpa-onnx>=1.10.42
+
+soundfile