Change the Singing Voice In Suno with Kits.ai

Bob Doyle Media
22 Mar 202417:02

TLDRIn this video, the host introduces Kits.ai, a platform for voice manipulation and music creation. They demonstrate how to use the site to convert one's voice to another, train a voice, blend vocal models, and remove vocals from songs. The host also explores AI mastering, vocal deharmonization, and the process of converting a voice to different musical instruments. They show the process of recording a voice, adjusting pitch levels, and applying audio effects like reverb and compression. The video also covers using the output from Sunno AI to create songs and then converting the singing voice using Kits.ai. The host concludes by discussing the potential of voice cloning and the use of RVC models, providing a comprehensive overview of the creative possibilities offered by AI in music production.

Takeaways

  • 🎧 The channel focuses on audio manipulation and showcases a site for voice cloning and AI music creation.
  • 🆓 Kits.ai offers many free functionalities for audio conversion and music creation.
  • 🔄 Conversion tools on Kits.ai allow for voice conversion, voice training, blending vocal models, vocal removal, and AI mastering.
  • 🎵 Users can select from a library of voices or clone their own using the platform's tools.
  • 🎤 The process includes recording a voice sample and adjusting pitch levels to match the chosen voice model.
  • 🎚 Advanced settings enable users to apply pre- and post-processing effects like reverb, compression, and delay.
  • 🔄 A demonstration of converting the host's voice to a female voice and then to a male voice is provided.
  • 🎼 Kits.ai can also separate vocal tracks from music, allowing for new vocals to be layered onto existing instrumentals.
  • 🎧 A tutorial on using Sunno AI to create a song and then process the vocals with Kits.ai is included.
  • 🎶 AI Mastering feature is tested, with options to choose presets or reference tracks for mastering audio.
  • 📚 The script mentions the possibility of uploading custom voice models, although the host's attempt with a Neville Goddard model was not compatible.
  • 🎹 Additionally, the platform allows users to use their voice to guide musical instruments, creating unique soundscapes.

Q & A

  • What is the main topic of the video?

    -The main topic of the video is about exploring the capabilities of a website called Kits.ai, which is focused on voice cloning and AI music. The video demonstrates how to use Kits.ai to change singing voices and manipulate audio tracks.

  • What are some of the free features available on Kits.ai?

    -Some of the free features on Kits.ai include the ability to convert one voice to another, train a voice, blend two vocal models, remove vocals from songs, and access to AI mastering and vocal deharmorization.

  • How does the process of changing a singing voice using Kits.ai start?

    -The process starts by selecting a voice from the available options in Kits.ai. Users can then upload an isolated vocal track, use a YouTube link, or record their own voice directly from the page to be converted.

  • What is the purpose of the advanced settings in Kits.ai?

    -The advanced settings in Kits.ai allow users to make adjustments to the pitch level to match the selected voice, as well as apply pre-processing and post-processing effects like compression, chorus, reverb, and delay to the audio.

  • How does Kits.ai handle pitch level discrepancies between the original voice and the chosen voice?

    -Kits.ai provides a pitch shift feature that allows users to adjust the pitch level of their voice to match the chosen voice. The system indicates when the pitch level is good or too much.

  • What is the role of AI mastering in the process?

    -AI mastering is a feature that can be used to enhance the quality of the final audio track. It can make the audio sound more professional by applying mastering techniques that improve the overall sound balance and clarity.

  • How can the output from Sunno AI be used with Kits.ai?

    -The output from Sunno AI, which is a song created from typed text, can be downloaded and the vocal track isolated using a vocal remover tool. This isolated vocal track can then be converted using Kits.ai to a different voice or style.

  • What is the purpose of the vocal remover tool?

    -The vocal remover tool is used to separate the vocal track from the music track in a song. This allows users to have the vocal and music tracks separately, which can then be used for further audio manipulation.

  • How does the video creator suggest using the AI mastering feature?

    -The video creator suggests using the AI mastering feature by uploading the final mixed audio track and choosing a preset or a reference track to guide the mastering process, aiming to improve the overall sound quality.

  • What is the significance of the 'My Voices' section in Kits.ai?

    -The 'My Voices' section in Kits.ai is where users can find, create, or clone their own voices. It serves as a library for personalized voice models that can be used for various audio projects.

  • How does the video demonstrate the use of RVC models in Kits.ai?

    -The video demonstrates the use of RVC (Recurrent Voice Cloning) models by attempting to upload a trained voice model for a thought leader named Neville Godard. However, it is mentioned that the model is not compatible, but the process shows how users can theoretically use their own trained voice models in Kits.ai.

Outlines

00:00

🎧 Voice Cloning and AI Music Exploration

The video begins with the host donning dorky headphones and diving into the world of audio manipulation. The focus is on a website that offers a range of tools for voice cloning and AI music creation. The host guides viewers through the process of using the site's features, such as converting one voice to another, blending vocal models, removing vocals from songs, and applying AI mastering. The host also demonstrates how to use the site's vocal de-harmonizer, which isolates the main vocal track from a layered harmony. The video provides a hands-on tutorial on selecting and converting voices, adjusting pitch levels, and applying audio effects like reverb and compression to achieve the desired sound.

05:00

🎵 Integrating Sunno AI for Song Creation

The host discusses how to integrate the site's voice conversion tools with Sunno AI, a platform for creating songs from text. After offering a tip to explore the styles of other users on Sunno AI for inspiration, the host creates a song using custom mode and random lyrics, then downloads it. The next step is to isolate the vocal track using a vocal remover tool, which allows for the separation of vocals and music. The host then demonstrates how to layer the converted vocal track onto the music track using Audacity, adjusting the volume levels and panning for a balanced mix. The segment concludes with a brief mention of the AI mastering feature, which is used to enhance the final song's audio quality.

10:01

🔊 AI Mastering and Voice Model Uploads

The host explores the AI mastering feature in more detail, comparing the original and mastered versions of a song to highlight the improvements in audio quality. The discussion then shifts to the 'My Voices' section, where the host expresses a desire to clone their own voice in the future. The video also covers the process of uploading voice models, specifically RVC (Reverse-Voice-Chain) models, which can be trained on a personal computer. The host attempts to upload a voice model for Neville Godard but encounters compatibility issues. The video concludes with a demonstration of using one's voice to guide musical instruments, experimenting with an overdriven guitar, drums, and bass guitar, and applying various audio effects to create unique sounds.

15:06

📚 Final Thoughts and Call to Action

The host summarizes the features of the website, emphasizing the creative potential of the tools available for free. They encourage viewers to explore the site and subscribe to the channel for more content on AI and creative applications. The host humorously warns that if viewers do not subscribe, they will be found and pursued, adding a playful tone to the call to action. The video ends with a reminder of the channel's focus on innovative uses of AI in creative projects.

Mindmap

Keywords

💡Voice Cloning

Voice cloning refers to the process of replicating a person's voice using AI technology. In the video, the host demonstrates how to use a website called Kits.ai to clone a voice and apply it to different audio tracks. This technology is significant as it allows for the creation of personalized voice models that can be used in various applications, such as virtual assistants or voiceovers.

💡AI Music

AI Music is a genre that involves the use of artificial intelligence to compose, perform, or produce music. The video showcases how the platform Sunno AI can generate songs in various styles from text input, and then how Kits.ai can be used to modify the singing voice of these AI-generated songs. This represents the intersection of music and technology, where AI is used to enhance creativity and produce novel sounds.

💡Bandon a Box

Bandon a Box is a software mentioned in the video that allows users to re-orchestrate music using MIDI instruments in various styles. It is used in conjunction with Sunno AI to modify the orchestration of the generated music. This tool exemplifies the use of technology to innovate in the music production process, offering musicians and producers new ways to experiment with sound.

💡Conversion Tools

Conversion tools in Kits.ai enable users to convert one voice to another, offering a range of functionalities such as voice training, blending vocal models, and vocal removal from songs. These tools are central to the video's demonstration, as they allow the host to change the singing voice of a song to different voices, showcasing the versatility of AI in voice manipulation.

💡AI Mastering

AI Mastering is a feature within Kits.ai that automatically optimizes the sound quality of a track, enhancing its professional polish. The host in the video uses this feature to improve the quality of an AI-generated song, highlighting the role of AI in post-production processes and how it can elevate the final output of a musical piece.

💡Harmonies

Harmonies refer to the simultaneous combination of tones sounded together in a chord, creating a richer and more complex sound. The video discusses a feature called 'Vocal Dearmoning' that can isolate the main vocal track from layers of harmonies. This is important for musicians looking to separate or modify individual vocal parts within a multi-layered recording.

💡Vocal Track

A vocal track is an individual recording of the singing or spoken voice within a song. The script describes how to isolate the vocal track from the music using tools like the vocal remover. This is a crucial step in the process of changing the singing voice in a song, as it allows for the original voice to be replaced with a cloned or different voice.

💡Pitch Level

Pitch level refers to the perceived frequency of a voice or sound, which determines how high or low it sounds. In the context of the video, adjusting the pitch level is necessary when converting a voice to match the pitch of the original singing voice, especially when there's a significant difference in the vocal range, such as changing a male voice to a female voice or vice versa.

💡Pre-processing and Post-processing Effects

Pre-processing and post-processing effects are audio techniques used to modify and enhance the quality of a sound recording. In the video, the host adds effects like reverb, chorus, and compression to the audio before and after it gets converted. These effects contribute to the final sound of the vocal track, giving it a more polished and professional finish.

💡Vocal Remover

A vocal remover is a tool that attempts to separate the vocal parts from the instrumental parts of a song. In the video, the host uses a vocal remover to isolate the vocals from the music track of an AI-generated song. This is a key step in the process of re-recording the song with a different voice using Kits.ai.

💡RVC Models

RVC Models refer to the Recurrent Variational Compression models, which are used for voice conversion and cloning. The video mentions that these models can be trained on any voice for free and then used in various applications. The host attempts to upload an RVC model of a specific voice for conversion purposes, demonstrating the potential for personalized voice cloning.

Highlights

The channel explores audio manipulation using AI with a focus on voice cloning and AI music.

Suno AI is used to generate songs from text in various styles.

Bandon a Box is mentioned for re-orchestrating music with MIDI instruments.

Kits.ai offers conversion tools to change a voice to another, train a voice, blend vocal models, and remove vocals from songs.

AI mastering and vocal deharmorization are features of Kits.ai that can enhance and isolate vocal tracks.

Users can create or clone voices in the 'My Voices' section of Kits.ai.

The process of recording and converting one's voice to another using Kits.ai is demonstrated.

Advanced settings in Kits.ai allow for pitch adjustment and audio effects like reverb and compression.

A demonstration of converting a male voice to a female voice using pitch shift is shown.

The video shows how to use Sunno AI to create a song and then isolate the vocal track for conversion.

Vocal remover tools can separate vocals from a music track to allow for new vocal overlays.

A step-by-step guide on how to mix converted vocals with original music using Audacity is provided.

The AI Mastering feature of Kits.ai is used to enhance the quality and loudness of a track.

The potential of using RVC (Reversed Vocoder) models for voice cloning in Kits.ai is discussed.

Uploading a custom voice model to Kits.ai for conversion is attempted, showcasing the process.

Kits.ai allows users to use their voice to guide musical instruments, demonstrated with guitar and drums.

An autotune feature and audio effects are experimented with to create unique vocal and instrument sounds.

The channel encourages viewers to explore AI and creative music, inviting them to subscribe for similar content.