[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-71042":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":16,"stars30d":17,"stars90d":16,"forks30d":16,"starsTrendScore":16,"compositeScore":18,"rankGlobal":10,"rankLanguage":10,"license":19,"archived":20,"fork":21,"defaultBranch":22,"hasWiki":21,"hasPages":21,"topics":23,"createdAt":10,"pushedAt":10,"updatedAt":31,"readmeContent":32,"aiSummary":33,"trendingCount":16,"starSnapshotCount":16,"syncStatus":17,"lastSyncTime":34,"discoverSource":35},71042,"VALL-E-X","Plachtaa\u002FVALL-E-X","Plachtaa","An open source implementation of Microsoft's VALL-E X zero-shot TTS model. Demo is available in https:\u002F\u002Fplachtaa.github.io\u002Fvallex\u002F","",null,"Python",7939,779,1,98,0,2,39.68,"MIT License",true,false,"master",[24,25,26,27,28,29,30],"emotional-speech","gpt","text-to-speech","transformer-architecture","tts","vall-e","voice-clone","2026-06-12 02:02:47","# VALL-E X: Multilingual Text-to-Speech Synthesis and Voice Cloning 🔊\n[![Discord](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-%235865F2.svg?style=for-the-badge&logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.gg\u002FqCBRmAnTxg)\n\u003Cbr>\nEnglish | [中文](README-ZH.md)\n\u003Cbr>\nAn open source implementation of Microsoft's [VALL-E X](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2303.03926) zero-shot TTS model.\u003Cbr>\n**We release our trained model to the public for research or application usage.**\n\n![vallex-framework](\u002Fimages\u002Fvallex_framework.jpg \"VALL-E X framework\")\n\nVALL-E X is an amazing multilingual text-to-speech (TTS) model proposed by Microsoft. While Microsoft initially publish in their research paper, they did not release any code or pretrained models. Recognizing the potential and value of this technology, our team took on the challenge to reproduce the results and train our own model. We are glad to share our trained VALL-E X model with the community, allowing everyone to experience the power next-generation TTS! 🎧\n\u003Cbr>\n\u003Cbr>\nMore details about the model are presented in [model card](.\u002Fmodel-card.md).\n\n## 📖 Quick Index\n* [🚀 Updates](#-updates)\n* [📢 Features](#-features)\n* [💻 Installation](#-installation)\n* [🎧 Demos](#-demos)\n* [🐍 Usage](#-usage-in-python)\n* [❓ FAQ](#-faq)\n* [🧠 TODO](#-todo)\n\n## 🚀 Updates\n**2023.09.10**\n- Added AR decoder batch decoding for more stable generation result.\n\n**2023.08.30**\n- Replaced EnCodec decoder with Vocos decoder, improved audio quality. (Thanks to [@v0xie](https:\u002F\u002Fgithub.com\u002Fv0xie))\n\n**2023.08.23**\n- Added long text generation.\n\n**2023.08.20**\n- Added [Chinese README](README-ZH.md).\n\n**2023.08.14**\n- Pretrained VALL-E X checkpoint is now released. Download it [here](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F10gdQWvP-K_e1undkvv0p2b7SU6I4Egyl\u002Fview?usp=sharing)\n\n## 💻 Installation\n### Install with pip, Python 3.10, CUDA 11.7 ~ 12.0, PyTorch 2.0+\n```commandline\ngit clone https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X.git\ncd VALL-E-X\npip install -r requirements.txt\n```\n\n> Note: If you want to make prompt, you need to install ffmpeg and add its folder to the environment variable PATH.\n\nWhen you run the program for the first time, it will automatically download the corresponding model. \n\nIf the download fails and reports an error, please follow the steps below to manually download the model.\n\n(Please pay attention to the capitalization of folders)\n\n1. Check whether there is a `checkpoints` folder in the installation directory. \nIf not, manually create a `checkpoints` folder (`.\u002Fcheckpoints\u002F`) in the installation directory.\n\n2. Check whether there is a `vallex-checkpoint.pt` file in the `checkpoints` folder. \nIf not, please manually download the `vallex-checkpoint.pt` file from [here](https:\u002F\u002Fhuggingface.co\u002FPlachta\u002FVALL-E-X\u002Fresolve\u002Fmain\u002Fvallex-checkpoint.pt) and put it in the `checkpoints` folder.\n\n3. Check whether there is a `whisper` folder in the installation directory. \nIf not, manually create a `whisper` folder (`.\u002Fwhisper\u002F`) in the installation directory.\n\n4. Check whether there is a `medium.pt` file in the `whisper` folder. \nIf not, please manually download the `medium.pt` file from [here](https:\u002F\u002Fopenaipublic.azureedge.net\u002Fmain\u002Fwhisper\u002Fmodels\u002F345ae4da62f9b3d59415adc60127b97c714f32e89e936602e85993674d08dcb1\u002Fmedium.pt) and put it in the `whisper` folder.\n\n##  🎧 Demos\nNot ready to set up the environment on your local machine just yet? No problem! We've got you covered with our online demos. You can try out VALL-E X directly on Hugging Face or Google Colab, experiencing the model's capabilities hassle-free!\n\u003Cbr>\n[![Open in Spaces](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F🤗-Open%20in%20Spaces-blue.svg)](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FPlachta\u002FVALL-E-X)\n[![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1yyD_sz531QntLKowMHo-XxorsFBCfKul?usp=sharing)\n\n\n## 📢 Features\n\nVALL-E X comes packed with cutting-edge functionalities:\n\n1. **Multilingual TTS**: Speak in three languages - English, Chinese, and Japanese - with natural and expressive speech synthesis.\n\n2. **Zero-shot Voice Cloning**: Enroll a short 3~10 seconds recording of an unseen speaker, and watch VALL-E X create personalized, high-quality speech that sounds just like them!\n\n\u003Cdetails>\n  \u003Csummary>\u003Ch5>see example\u003C\u002Fh5>\u003C\u002Fsummary>\n\n[prompt.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Fa7baa51d-a53a-41cc-a03d-6970f25fcca7)\n\n\n[output.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Fb895601a-d126-4138-beff-061aabdc7985)\n\n\u003C\u002Fdetails>\n\n3. **Speech Emotion Control**: Experience the power of emotions! VALL-E X can synthesize speech with the same emotion as the acoustic prompt provided, adding an extra layer of expressiveness to your audio.\n\n\u003Cdetails>\n  \u003Csummary>\u003Ch5>see example\u003C\u002Fh5>\u003C\u002Fsummary>\n\nhttps:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002F56fa9988-925e-4757-82c5-83ecb0df6266\n\n\nhttps:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002F699c47a3-d502-4801-8364-bd89bcc0b8f1\n\n\u003C\u002Fdetails>\n\n4. **Zero-shot Cross-Lingual Speech Synthesis**: Take monolingual speakers on a linguistic journey! VALL-E X can produce personalized speech in another language without compromising on fluency or accent. Below is a Japanese speaker talk in Chinese & English. 🇯🇵 🗣\n\n\u003Cdetails>\n  \u003Csummary>\u003Ch5>see example\u003C\u002Fh5>\u003C\u002Fsummary>\n\n[jp-prompt.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Fea6e2ee4-139a-41b4-837e-0bd04dda6e19)\n\n\n[en-output.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Fdb8f9782-923f-425e-ba94-e8c1bd48f207)\n\n\n[zh-output.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002F15829d79-e448-44d3-8965-fafa7a3f8c28)\n\n\u003C\u002Fdetails>\n\n5. **Accent Control**: Get creative with accents! VALL-E X allows you to experiment with different accents, like speaking Chinese with an English accent or vice versa. 🇨🇳 💬\n\n\u003Cdetails>\n  \u003Csummary>\u003Ch5>see example\u003C\u002Fh5>\u003C\u002Fsummary>\n\n[en-prompt.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Ff688d7f6-70ef-46ec-b1cc-355c31e78b3b)\n\n\n[zh-accent-output.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Fbe59c7ca-b45b-44ca-a30d-4d800c950ccc)\n\n\n[en-accent-output.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002F8b4f4f9b-f299-4ea4-a548-137437b71738)\n\n\u003C\u002Fdetails>\n\n6. **Acoustic Environment Maintenance**: No need for perfectly clean audio prompts! VALL-E X adapts to the acoustic environment of the input, making speech generation feel natural and immersive.\n\n\u003Cdetails>\n  \u003Csummary>\u003Ch5>see example\u003C\u002Fh5>\u003C\u002Fsummary>\n\n[noise-prompt.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002F68986d88-abd0-4d1d-96e4-4f893eb9259e)\n\n\n[noise-output.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002F96c4c612-4516-4683-8804-501b70938608)\n\n\u003C\u002Fdetails>\n\n\nExplore our [demo page](https:\u002F\u002Fplachtaa.github.io\u002F) for a lot more examples!\n\n## 🐍 Usage in Python\n\n\u003Cdetails open>\n  \u003Csummary>\u003Ch3>🪑 Basics\u003C\u002Fh3>\u003C\u002Fsummary>\n\n```python\nfrom utils.generation import SAMPLE_RATE, generate_audio, preload_models\nfrom scipy.io.wavfile import write as write_wav\nfrom IPython.display import Audio\n\n# download and load all models\npreload_models()\n\n# generate audio from text\ntext_prompt = \"\"\"\nHello, my name is Nose. And uh, and I like hamburger. Hahaha... But I also have other interests such as playing tactic toast.\n\"\"\"\naudio_array = generate_audio(text_prompt)\n\n# save audio to disk\nwrite_wav(\"vallex_generation.wav\", SAMPLE_RATE, audio_array)\n\n# play text in notebook\nAudio(audio_array, rate=SAMPLE_RATE)\n```\n\n[hamburger.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002F578d7bbe-cda9-483e-898c-29646edc8f2e)\n\n\u003C\u002Fdetails>\n\n\u003Cdetails open>\n  \u003Csummary>\u003Ch3>🌎 Foreign Language\u003C\u002Fh3>\u003C\u002Fsummary>\n\u003Cbr>\nThis VALL-E X implementation also supports Chinese and Japanese. All three languages have equally awesome performance!\n\u003Cbr>\n\n```python\n\ntext_prompt = \"\"\"\n    チュソクは私のお気に入りの祭りです。 私は数日間休んで、友人や家族との時間を過ごすことができます。\n\"\"\"\naudio_array = generate_audio(text_prompt)\n```\n\n[vallex_japanese.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Fee57a688-3e83-4be5-b0fe-019d16eec51c)\n\n*Note: VALL-E X controls accent perfectly even when synthesizing code-switch text. However, you need to manually denote language of respective sentences (since our g2p tool is rule-base)*\n```python\ntext_prompt = \"\"\"\n    [EN]The Thirty Years' War was a devastating conflict that had a profound impact on Europe.[EN]\n    [ZH]这是历史的开始。 如果您想听更多，请继续。[ZH]\n\"\"\"\naudio_array = generate_audio(text_prompt, language='mix')\n```\n\n[vallex_codeswitch.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Fd8667abf-bd08-499f-a383-a861d852f98a)\n\n\u003C\u002Fdetails>\n\n\u003Cdetails open>\n\u003Csummary>\u003Ch3>📼 Voice Presets\u003C\u002Fh3>\u003C\u002Fsummary>\n  \nVALL-E X provides tens of speaker voices which you can directly used for inference! Browse all voices in the [code](\u002Fpresets)\n\n> VALL-E X tries to match the tone, pitch, emotion and prosody of a given preset. The model also attempts to preserve music, ambient noise, etc.\n\n```python\ntext_prompt = \"\"\"\nI am an innocent boy with a smoky voice. It is a great honor for me to speak at the United Nations today.\n\"\"\"\naudio_array = generate_audio(text_prompt, prompt=\"dingzhen\")\n```\n\n[smoky.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Fd3f55732-b1cd-420f-87d6-eab60db14dc5)\n\n\u003C\u002Fdetails>\n\n\u003Cdetails open>\n\u003Csummary>\u003Ch3>🎙Voice Cloning\u003C\u002Fh3>\u003C\u002Fsummary>\n  \nVALL-E X supports voice cloning! You can make a voice prompt with any person, character or even your own voice, and use it like other voice presets.\u003Cbr>\nTo make a voice prompt, you need to provide a speech of 3~10 seconds long, as well as the transcript of the speech. \nYou can also leave the transcript blank to let the [Whisper](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fwhisper) model to generate the transcript.\n> VALL-E X tries to match the tone, pitch, emotion and prosody of a given prompt. The model also attempts to preserve music, ambient noise, etc.\n\n```python\nfrom utils.prompt_making import make_prompt\n\n### Use given transcript\nmake_prompt(name=\"paimon\", audio_prompt_path=\"paimon_prompt.wav\",\n                transcript=\"Just, what was that? Paimon thought we were gonna get eaten.\")\n\n### Alternatively, use whisper\nmake_prompt(name=\"paimon\", audio_prompt_path=\"paimon_prompt.wav\")\n```\nNow let's try out the prompt we've just made!\n```python\nfrom utils.generation import SAMPLE_RATE, generate_audio, preload_models\nfrom scipy.io.wavfile import write as write_wav\n\n# download and load all models\npreload_models()\n\ntext_prompt = \"\"\"\nHey, Traveler, Listen to this, This machine has taken my voice, and now it can talk just like me!\n\"\"\"\naudio_array = generate_audio(text_prompt, prompt=\"paimon\")\n\nwrite_wav(\"paimon_cloned.wav\", SAMPLE_RATE, audio_array)\n\n```\n\n[paimon_prompt.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002Fe7922859-9d12-4e2a-8651-e156e4280311)\n\n\n[paimon_cloned.webm](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fassets\u002F112609742\u002F60d3b7e9-5ead-4024-b499-a897ce5f3d5e)\n\n\n\u003C\u002Fdetails>\n\n\n\u003Cdetails open>\n\u003Csummary>\u003Ch3>🎢User Interface\u003C\u002Fh3>\u003C\u002Fsummary>\n\nNot comfortable with codes? No problem! We've also created a user-friendly graphical interface for VALL-E X. It allows you to interact with the model effortlessly, making voice cloning and multilingual speech synthesis a breeze.\n\u003Cbr>\nYou can launch the UI by the following command:\n```commandline\npython -X utf8 launch-ui.py\n```\n\u003C\u002Fdetails>\n\n## 🛠️ Hardware and Inference Speed\n\nVALL-E X works well on both CPU and GPU (`pytorch 2.0+`, CUDA 11.7 and CUDA 12.0).\n\nA GPU VRAM of 6GB is enough for running VALL-E X without offloading.\n\n## ⚙️ Details\n\nVALL-E X is similar to [Bark](https:\u002F\u002Fgithub.com\u002Fsuno-ai\u002Fbark), [VALL-E](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.02111) and [AudioLM](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.03143), which generates audio in GPT-style by predicting audio tokens quantized by [EnCodec](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fencodec).\n\u003Cbr>\nComparing to [Bark](https:\u002F\u002Fgithub.com\u002Fsuno-ai\u002Fbark):\n- ✔ **Light-weighted**: 3️⃣ ✖ smaller,\n- ✔ **Efficient**: 4️⃣ ✖ faster, \n- ✔ **Better quality on Chinese & Japanese**\n- ✔ **Cross-lingual speech without foreign accent**\n- ✔ **Easy voice-cloning**\n- ❌ **Less languages**\n- ❌ **No special tokens for music \u002F sound effects**\n\n### Supported Languages\n\n| Language | Status |\n| --- | :---: |\n| English (en) | ✅ |\n| Japanese (ja) | ✅ |\n| Chinese, simplified (zh) | ✅ |\n\n## ❓ FAQ\n\n#### Where is code for training?\n* [lifeiteng's vall-e](https:\u002F\u002Fgithub.com\u002Flifeiteng\u002Fvall-e) has almost everything. There is no plan to release our training code because there is no difference between lifeiteng's implementation.\n\n#### Where can I download the model checkpoint?\n* We use `wget` to download the model to directory `.\u002Fcheckpoints\u002F` when you run the program for the first time.\n* If the download fails on the first run, please manually download from [this link](https:\u002F\u002Fhuggingface.co\u002FPlachta\u002FVALL-E-X\u002Fresolve\u002Fmain\u002Fvallex-checkpoint.pt), and put the file under directory `.\u002Fcheckpoints\u002F`.\n\n#### How much VRAM do I need?\n* 6GB GPU VRAM - Almost all NVIDIA GPUs satisfy the requirement.\n\n#### Why the model fails to generate long text?\n* Transformer's computation complexity increases quadratically while the sequence length increases. Hence, all training \nare kept under 22 seconds. Please make sure the total length of audio prompt and generated audio is less than 22 seconds \nto ensure acceptable performance. \n\n\n#### MORE TO BE ADDED...\n\n## 🧠 TODO\n- [x] Add Chinese README\n- [x] Long text generation\n- [x] Replace Encodec decoder with Vocos decoder\n- [ ] Fine-tuning for better voice adaptation\n- [ ] `.bat` scripts for non-python users\n- [ ] To be added...\n\n## 🙏 Appreciation\n- [VALL-E X paper](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2303.03926) for the brilliant idea\n- [lifeiteng's vall-e](https:\u002F\u002Fgithub.com\u002Flifeiteng\u002Fvall-e) for related training code\n- [bark](https:\u002F\u002Fgithub.com\u002Fsuno-ai\u002Fbark) for the amazing pioneering work in neuro-codec TTS model\n\n## ⭐️ Show Your Support\n\nIf you find VALL-E X interesting and useful, give us a star on GitHub! ⭐️ It encourages us to keep improving the model and adding exciting features.\n\n## 📜 License\n\nVALL-E X is licensed under the [MIT License](.\u002FLICENSE).\n\n---\n\nHave questions or need assistance? Feel free to [open an issue](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X\u002Fissues\u002Fnew) or join our [Discord](https:\u002F\u002Fdiscord.gg\u002FqCBRmAnTxg)\n\nHappy voice cloning! 🎤\n","VALL-E X 是一个开源的多语言文本转语音（TTS）和声音克隆模型，基于微软提出的零样本TTS模型。该项目使用Python开发，具备情感语音合成、GPT集成及Transformer架构等核心技术特点，能够实现高质量的语音合成与个性化声音复制。适用于需要自然流畅语音输出的应用场景，如虚拟助手、有声读物制作或游戏配音等。项目已发布训练好的模型供研究和应用使用，并持续更新以提升音频质量和生成稳定性。","2026-06-11 03:35:37","high_star"]