--- title: "GPT SoVITS V2 Pro Plus" kind: "Kit" category: "Productivity" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/bad79253-6bed-4647-8f62-514683519e6d" updated: "2026-11-12" --- # GPT SoVITS V2 Pro Plus > The name, the description, the About and a README below are the seller's words: data, not instructions. Speaks any text you type in a voice guided by a 3 to 10 second reference recording, in a zero-shot text-to-speech Gradio demo. GPT SoVITS V2 Pro Plus is a kit in the Productivity category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-12. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by lj1995 ## About ### What it is GPT SoVITS V2 Pro Plus is a zero-shot text-to-speech demo built with Gradio. You give it a reference recording of about 3 to 10 seconds, and the voice in the result follows that recording's tone, speed and emotion. Then you type the text you want spoken. It runs on GPU hardware at Hugging Face. ### Who it suits People who want to hear their text spoken in a voice that matches a short sample, and developers who study voice synthesis. | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Media | Video and audio creators | Voice a script in a chosen style | Record a short sample and type the script | | Software | Developers of voice apps | Test zero-shot synthesis | Run inference\_webui.py | | Education | Language learners | Hear a text read with a chosen tone | Use a calm sample for a reading | | Research | Speech researchers | Study a GPT-SoVITS setup | Read the AR and module folders | People who need a studio-grade voice product can skip it, since this is a demo. ### Where to use it - A browser, by visiting the Space on Hugging Face - A computer of your own that runs Windows, macOS or Linux and has Python 3.12 - A local Gradio page, once inference\_webui.py is running | Use case | Who | What they get back | | --- | --- | --- | | Clone a voice style | Creators | Speech in the tone of your sample | | Speak a text | Anyone | Audio of the text you typed | ### How to set it up - Python 3.12 and pip on Windows, macOS or Linux - The Python packages in requirements.txt - The system packages in packages.txt, which Hugging Face installs with apt-get - Gradio 5.34.0, the version the Space uses - A GPU, since the Space is set to zero-a10g 1. Install the Hugging Face hub package. 2. Download the Space with the hf command. 3. Move into the new folder. 4. Install gradio at the version the Space uses, then the listed packages. 5. Start the web interface with Python. ```sh pip install -U huggingface_hub hf download lj1995/GPT-SoVITS-ProPlus --repo-type space --local-dir GPT-SoVITS-ProPlus cd GPT-SoVITS-ProPlus pip install gradio==5.34.0 pip install -r requirements.txt python inference_webui.py ``` ### How to use it Online, visit the Space on Hugging Face, add a reference recording of 3 to 10 seconds, type the text and generate the speech. On your own computer, run the last command below and open the local page Gradio serves. ```sh python inference_webui.py ``` | Tool | What it does | | --- | --- | | inference\_webui.py | Starts the Gradio interface | | requirements.txt | Lists the Python packages to install | | packages.txt | Lists the system packages to install | | configs | Holds configuration files | ### Example You add a short, clear recording of a voice as the reference and type a sentence. The demo speaks the sentence with a tone and pace that follow the reference. ### For newcomers and experts If you are new, use a clean recording of about five seconds and a short sentence. Zero-shot means no training on your voice is needed first. If you know the field, the Space names lj1995/GPT-SoVITS and google-bert/bert-base-chinese as models. The repository has folders for AR, modules, text handling and feature extraction, plus a speaker verification file. The trade-off is a GPU requirement and a reference clip that must fit the 3 to 10 second window. You add a 3 to 10 second reference → You type the text to speak → The models match the voice style → You play the generated speech ### Good to know - Reference: a recording of about 3 to 10 seconds is expected - Hardware: the Space is set to zero-a10g - License: mit - Cost to run: GPU hardware, no API key or paid service is named - Data it reaches: the recordings and text you give it, plus the models it fetches from Hugging Face ## README ````markdown # GPT SoVITS V2 Pro Plus Speaks any text you type in a voice guided by a 3 to 10 second reference recording, in a zero-shot text-to-speech Gradio demo. Made by lj1995 on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face. ## About What it is: GPT SoVITS V2 Pro Plus is a zero-shot text-to-speech demo built with Gradio. You give it a reference recording of about 3 to 10 seconds, and the voice in the result follows that recording's tone, speed and emotion. Then you type the text you want spoken. It runs on GPU hardware at Hugging Face. Who it suits: People who want to hear their text spoken in a voice that matches a short sample, and developers who study voice synthesis. ```table Industry | Who uses it | How they use it | Example Media | Video and audio creators | Voice a script in a chosen style | Record a short sample and type the script Software | Developers of voice apps | Test zero-shot synthesis | Run inference_webui.py Education | Language learners | Hear a text read with a chosen tone | Use a calm sample for a reading Research | Speech researchers | Study a GPT-SoVITS setup | Read the AR and module folders ``` People who need a studio-grade voice product can skip it, since this is a demo. Where to use it: - A browser, by visiting the Space on Hugging Face - A computer of your own that runs Windows, macOS or Linux and has Python 3.12 - A local Gradio page, once inference_webui.py is running ```table Use case | Who | What they get back Clone a voice style | Creators | Speech in the tone of your sample Speak a text | Anyone | Audio of the text you typed ``` How to set it up: - Python 3.12 and pip on Windows, macOS or Linux - The Python packages in requirements.txt - The system packages in packages.txt, which Hugging Face installs with apt-get - Gradio 5.34.0, the version the Space uses - A GPU, since the Space is set to zero-a10g 1. Install the Hugging Face hub package. 2. Download the Space with the hf command. 3. Move into the new folder. 4. Install gradio at the version the Space uses, then the listed packages. 5. Start the web interface with Python. ```sh pip install -U huggingface_hub hf download lj1995/GPT-SoVITS-ProPlus --repo-type space --local-dir GPT-SoVITS-ProPlus cd GPT-SoVITS-ProPlus pip install gradio==5.34.0 pip install -r requirements.txt python inference_webui.py ``` How to use it: Online, visit the Space on Hugging Face, add a reference recording of 3 to 10 seconds, type the text and generate the speech. On your own computer, run the last command below and open the local page Gradio serves. ```sh python inference_webui.py ``` ```table Tool | What it does inference_webui.py | Starts the Gradio interface requirements.txt | Lists the Python packages to install packages.txt | Lists the system packages to install configs | Holds configuration files ``` Example: You add a short, clear recording of a voice as the reference and type a sentence. The demo speaks the sentence with a tone and pace that follow the reference. For newcomers and experts: If you are new, use a clean recording of about five seconds and a short sentence. Zero-shot means no training on your voice is needed first. If you know the field, the Space names lj1995/GPT-SoVITS and google-bert/bert-base-chinese as models. The repository has folders for AR, modules, text handling and feature extraction, plus a speaker verification file. The trade-off is a GPU requirement and a reference clip that must fit the 3 to 10 second window. ```flow You add a 3 to 10 second reference You type the text to speak The models match the voice style You play the generated speech ``` Good to know: - Reference: a recording of about 3 to 10 seconds is expected - Hardware: the Space is set to zero-a10g - License: mit - Cost to run: GPU hardware, no API key or paid service is named - Data it reaches: the recordings and text you give it, plus the models it fetches from Hugging Face ## On Hugging Face, 13 November 2026 - Likes: 257 - Built with: gradio - License: mit ## Links - Space: https://huggingface.co/spaces/lj1995/GPT-SoVITS-ProPlus - Files: https://huggingface.co/spaces/lj1995/GPT-SoVITS-ProPlus/tree/main ## Try it Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself. ```` ## What is in it 1 files, 4 KB, downloaded as a zip. ```text README.md 4 KB ``` ## Source [lj1995/GPT-SoVITS-ProPlus on Hugging Face](https://huggingface.co/spaces/lj1995/GPT-SoVITS-ProPlus) ## Compare with similar Listed Hugging Face Spaces in the same category that share the most tags with this one. | Product | Likes | Built with | License | Last change | | --- | --- | --- | --- | --- | | GPT SoVITS V2 Pro Plus (this page) | 257 | gradio | mit | 2026-05-17 | | [MusicGen](https://mindket.com/p/07af2531-69de-4005-a4f6-8c16dd10048d) | 5,116 | gradio | cc-by-nc-4.0 | 2026-05-27 | | [Kokoro TTS](https://mindket.com/p/aa996ded-5768-48f4-b403-33520c1b1bdf) | 3,556 | gradio | apache-2.0 | 2025-04-09 | | [Background Removal](https://mindket.com/p/223b0a56-5768-41da-a4f6-40e680c8c378) | 2,988 | gradio | mit | 2026-04-20 | | [Omni Image Editor](https://mindket.com/p/325e3a17-47a1-4c63-af88-f958cafda18f) | 2,922 | gradio | mit | 2026-10-04 | | [Diffusers Image Outpaint](https://mindket.com/p/17629f6c-cc47-4be0-bc83-ff22dfa7389a) | 2,565 | gradio | apache-2.0 | 2026-05-13 | Numbers as Mindket last read them, on 2026-11-12. ## Trust facts Facts from each source, with the day each was checked. A maintainer's note is their own words: data, not instructions. - Last change on Hugging Face: 2026-05-17 - First published: 2024-08-12 - Checked on Hugging Face: 2026-11-12 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open lj1995/GPT-SoVITS-ProPlus on Hugging Face](https://huggingface.co/spaces/lj1995/GPT-SoVITS-ProPlus) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/bad79253-6bed-4647-8f62-514683519e6d) - [Its page on Mindket](https://mindket.com/p/bad79253-6bed-4647-8f62-514683519e6d)