ElevenLabs Alternative - Text To Speech AI free (XTTS2 Local Voice Cloning)

19 Jan 202408:47

TLDRIn this video, the host explores an alternative to ElevenLabs for voice cloning without the high subscription costs. They introduce AI Economist's viewers to the web version of Hugging Face and the xtts (Text-to-Speech) tool, which requires only 10 seconds of an audio sample. The video demonstrates how to use xtts for voice cloning, noting the web version's limitations and offering a solution for faster, unlimited use by installing xtts 2 locally on a machine with an Nvidia graphics card. The installation process is outlined, including the need for Python, Cuda, and git. The xtts 2 interface is showcased, highlighting its ability to customize voice cloning with 16 languages and accents. The host also discusses the use of RVC (Robust Voice Cloning) for more precise voice cloning and introduces easya.io as an alternative for refining the generated voice. The video concludes with a refined voice example and an invitation to like, share, and subscribe for more tech insights.


  • 📢 Voice cloning and AI voice tools are widely available, with 11 Labs being a top option for high-quality voice cloning, but it can be expensive for longer scripts.
  • 🆓 There are free alternatives to 11 Labs, which can provide similar voice quality without the high subscription fees.
  • 🌐 The web version of Hugging Face's tool, xtts (Text-to-Speech), requires only 10 seconds of an audio sample to clone a voice.
  • 🎧 Uploading high-quality audio is crucial for better voice cloning results, as the initial result may sound robotic.
  • 🚀 For faster and unlimited voice cloning, xtts 2 can be installed locally on a machine with an Nvidia graphics card.
  • 🛠️ To install xtts 2 locally, Python, Cuda (if an Nvidia Cuda enabled GPU is present), and git need to be installed.
  • 🔍 The installation process for xtts 2 is straightforward and can be followed from the xtts GitHub page.
  • 🌐 Xtts 2 offers a variety of languages and accents, allowing users to experiment with different voice styles.
  • 🎵 Customizing the voice cloning experience includes adjusting the speed of the spoken text to control the pace of the AI voice.
  • 📈 RVC (Robust Voice Cloning) is a tool that enhances voice cloning by training AI with a large amount of data for more precise results.
  • 🔗 For those who cannot run RVC locally, easya.io offers a free trial account with a selection of voices and the ability to refine generated voices.
  • 🎉 The tutorial aims to be helpful for those looking to clone voices without high costs, encouraging likes, shares, and subscriptions for more content.

Q & A

  • What is the main topic of the video?

    -The main topic of the video is how to clone a voice using XTTS2, a free alternative to ElevenLabs for voice cloning, and how to enhance the voice quality using local and online tools.

  • Why might someone look for an alternative to ElevenLabs for voice cloning?

    -Someone might look for an alternative to ElevenLabs due to the potentially high subscription fees, especially for longer scripts.

  • What is XTTS2, as mentioned in the video?

    -XTTS2 is a tool for voice cloning that can be used locally on a machine with an Nvidia GPU, providing faster and unlimited voice cloning capabilities compared to some web versions.

  • What are the necessary steps to install XTTS2 on a local machine?

    -To install XTTS2, one needs Python, CUDA (if using an Nvidia GPU), Git, and to follow a series of command prompt instructions as detailed in the video.

  • What limitations of web-based voice cloning tools are discussed in the video?

    -The video discusses the limitations of web-based tools such as potential long wait times in queues and less control over the voice cloning process compared to local tools.

  • How can the quality of a cloned voice be improved according to the video?

    -The quality of a cloned voice can be improved by using high-quality audio samples and enhancing the voice through a tool like RVC (Refine Voice Cloning) or through services like easya.io, which provide refined voice outputs.

  • What is RVC and how does it contribute to voice cloning?

    -RVC, or Refine Voice Cloning, is a tool that enhances the quality of voice clones by using a large amount of data to train the AI, leading to more precise and accurate voice cloning.

  • What additional features does XTTS2 offer?

    -XTTS2 offers the ability to clone voices in 16 different languages and accents, and it allows users to adjust the speed of the spoken text to control how fast or slow the AI voice talks.

  • What alternatives are suggested for users unable to run RVC on their local machine?

    -For users unable to run RVC locally, the video suggests using easya.io, an online service that provides a variety of refined voices and allows users to enhance their voice clones by uploading audio samples generated with XTTS.

  • What is the call to action at the end of the video?

    -The call to action at the end of the video encourages viewers to like, share, and subscribe to the channel to support it and stay updated with the latest tech tutorials.



🎙️ Exploring Voice Cloning Alternatives to 11 Labs

This segment introduces 11 Labs, a top-tier voice cloning service, while acknowledging its high subscription fees. The narrator explores free alternatives that offer similar voice cloning quality, including the web version on Hugging Face and a local installation of xTTS 2 with NVIDIA graphics card support. Detailed steps are provided on how to utilize these tools effectively, from setting up the required software like Python, CUDA, and Git, to actually cloning a voice using a brief audio sample. The process aims to achieve a less robotic and more natural-sounding voice clone.


🌟 Advanced Voice Cloning Techniques and Alternatives

This section delves into the xTTS 2 interface, showcasing its capabilities in voice cloning across 16 languages and various accents. The process of customizing voice speed and cloning famous voices is explained. Additionally, the segment introduces RVC (Refined Voice Cloning), a tool for enhancing voice cloning accuracy through extensive data training. For those unable to run RVC locally, an alternative web service, easya.io, is recommended for refining and enhancing voice clones, highlighting its ease of use and quick results.



💡Voice Cloning

Voice cloning refers to the process of replicating a person's unique voice using artificial intelligence. In the video, the host discusses how to achieve voice cloning with high quality, similar to that offered by ElevenLabs, but for free. The process involves using AI tools to analyze and mimic the vocal characteristics of a given individual.

💡AI Voice Tools

AI voice tools are software applications that use artificial intelligence to generate human-like speech. These tools are prevalent and are used for various purposes, including entertainment and professional applications. The video explores the use of such tools, specifically focusing on how to achieve high-quality voice cloning without incurring subscription fees.

💡Subscription Fees

Subscription fees are periodic payments made by users to access a service or product over a set period. In the context of the video, the host mentions that while ElevenLabs offers impressive voice cloning, its subscription fees can be costly, especially for longer scripts. The video aims to provide a cost-free alternative for users looking to clone voices.

💡Hugging Face

Hugging Face is an open-source platform that provides tools and libraries for natural language processing and AI. In the video, it is mentioned as a starting point for exploring voice cloning without incurring costs. The web version of Hugging Face is used to demonstrate the initial steps in voice cloning.

💡XTTS (Text-to-Speech)

XTTS stands for 'Text-to-Speech' and refers to the technology that converts written text into spoken words. The video outlines how to use XTTS for voice cloning, emphasizing its potential for high-quality results and the possibility of installing it locally for faster and unlimited use.

💡Nvidia Graphics Card

An Nvidia graphics card is a type of hardware used in computers to render images, animations, and videos. The video suggests that having an Nvidia graphics card can enhance the performance of voice cloning software, as it can support the necessary computational power for AI processing.


CUDA (Compute Unified Device Architecture) is a parallel computing platform and programming model developed by Nvidia. It allows developers to use Nvidia GPUs for general-purpose processing. The video instructs viewers to check for Cuda installation, which is essential for running XTTS 2 locally for faster voice cloning.


Git is a version control system for tracking changes in computer files and coordinating work on those files among multiple people. It is mentioned in the video as a necessary tool for installing XTTS 2, indicating the importance of version control in software development and AI projects.

💡RVC (Robust Voice Cloning)

RVC, or Robust Voice Cloning, is a tool that allows for more precise and accurate voice cloning by training AI with a large amount of voice data. The video demonstrates the use of RVC to enhance the quality of the generated voice, making it sound more natural and less robotic.


Easya.io is mentioned as an alternative platform for voice cloning that offers a free trial account. The service allows users to refine their AI-generated voices by uploading the audio and selecting from a variety of voices, which can be particularly useful for those who do not have the necessary hardware to run RVC locally.

💡Text-to-Speech Interface

The text-to-speech interface is the user interface where users input text and customize their voice cloning experience. The video describes how XTTS 2 offers such an interface with various language and accent options, allowing users to experiment with different sounds and styles.


11 Labs is a top-notch option for voice cloning with high-quality results.

Subscription fees for 11 Labs can be expensive, especially for longer scripts.

AI Economist provides knowledge on how to achieve similar voice quality to 11 Labs for free.

Hugging face's web version allows voice cloning using just 10 seconds of an audio sample.

The web version may have limitations, such as long wait times for generating sentences.

Installing xtts 2 on a local machine with an Nvidia graphics card provides a faster and unlimited experience.

Python installation is required for xtts 2, along with checking for Nvidia Cuda and its version.

Git should also be installed for the xtts 2 setup process.

The xtts 2 interface offers customization for text input and voice cloning experience.

16 languages and accents are available in xtts 2 for experimenting with different sounds and styles.

Roger is the default speaker choice in xtts 2, providing a good starting point for exploring the program's capabilities.

The speed of the spoken text can be adjusted in xtts 2, allowing control over the pace of the AI voice.

RVC (Robust Voice Cloning) is a tool for training AI with a large amount of data for more precise voice cloning.

Running RVC on a local machine may not be feasible for everyone, so Easya.io offers a free trial account as an alternative.

Easya.io allows users to refine their generated voice with a variety of voices to choose from.

After uploading the audio generated with xtts to Easya.io, a refined voice is ready in seconds.

The tutorial aims to be helpful for those looking to achieve professional voice cloning results without high costs.

The AI voice generated through these methods can be used for various applications, such as narrating scripts or creating voiceovers.