AI Music Kitbash! MIDI to Vocal Harmonies!

In this video, the creator embarks on a unique musical journey using AI tools to generate a complete track with sung vocals and harmonies, without the need for real singers or instruments. The process begins with Band-in-a-Box, a music accompaniment program, to create a song with a melody track. The melody is then transformed into sung lyrics using Ace Studio, an AI software that generates vocals from MIDI files. The creator explores various voice models and styles, eventually opting for a more powerful female voice for the lead vocals. Harmonies are added using Band-in-a-Box's real-time generation feature. The project concludes with the integration of a vocal jazz solo created in Band-in-a-Box and further refined in Adobe Audition for a polished finish. The video serves as an educational demonstration of how AI can be creatively utilized in music production, showcasing the potential and limitations of current AI music generation tools.

Takeaways

• 🎼 The video demonstrates how to create a music track with sung vocals and harmonies using AI tools, without real singers or instruments.
• 🤖 AI music generation is a controversial topic with strong opinions on its use and the quality of the tools available.
• 🔍 The creator uses Band-in-a-Box, a music accompaniment program, to generate a song and assign musical styles to chords.
• 🎶 The process involves changing the style of the song to a Samba groove and filtering for real tracks with the most realistic sounds.
• 📁 The melody track is exported as a MIDI file, and the music track is exported as a WAV file for further editing.
• 🎙️ Ace Studio is used to convert the melody line into sung lyrics that can be mixed with the audio track.
• 👩‍🎤 Different voice models in Ace Studio are optimized for various languages, with a subset being more suitable for English lyrics.
• ➕ Vocal harmonies are created by generating a Harmony track in Band-in-a-Box and converting it to a Melody track.
• 🔄 Vocal tracks can be doubled and shifted in octaves to add depth and a fuller sound to the music.
• 🌐 KitSplit is used to convert the lead vocal track into different voice styles, avoiding copyright or licensing issues.
• 🎵 Adobe Audition is suggested for final mixing, allowing for more control over effects and track levels.
• 📈 The video serves as an educational demonstration on how to use AI tools for music creation, encouraging creativity within the process.

Q & A

What is the main topic of the video?

The video is about creating a music track with sung vocals and harmonies using AI tools, without the involvement of real singers or instruments.

What is Band in a Box used for in the video?

Band in a Box is a music accompaniment program used to create a finished song by plugging in chords and assigning musical styles to them.

How does the AI software Ace Studio contribute to the music creation process?

Ace Studio is used to convert the melody line into sung lyrics that can be mixed with the audio track. It also allows for the layering of vocals and the application of different voice models for various vocal tracks.

What is the role of the song 'So Nice' in the video?

The song 'So Nice' is used as a recognizable and harmonically interesting piece to demonstrate how the AI tools can work together to create a music track with vocals and harmonies.

How does the video address copyright concerns?

The video acknowledges potential copyright issues and states that the use of the song 'So Nice' is for educational purposes only to demonstrate the capabilities of the AI tools.

What is the process of generating harmonies in Band in a Box?

To generate harmonies, a specific track is selected, and the Melody Harmony feature is used to preview and select the desired harmony. The harmony can then be converted to a Melody track and exported as a MIDI file.

How are vocals added to the melody track in Ace Studio?

Vocals are added by selecting a voice model from the list and dragging it onto the singer track. Lyrics can then be pasted into the software to align with the melody.

What is the purpose of creating vocal doubles in Ace Studio?

Creating vocal doubles helps to add depth to the vocals by generating slightly different pitch and timing variations of the original vocal track, which can then be panned to the left or right in the mix.

How does the video demonstrate the use of different AI voice models?

The video shows how to export a singer track from Ace Studio and import it into KitSplit, where different voice models are tested to find the most suitable one for the lead vocals.

What is the final step in the music creation process shown in the video?

The final step is to export all the generated tracks as separate stems and import them into a digital audio workstation (DAW) like Adobe Audition for further mixing, adding effects, and fine-tuning.

How does the video ensure the vocals are realistic and not overly processed?

The video emphasizes using voice models trained to optimize various languages and ensuring that the vocal samples are within the range of the models to maintain a natural sound.

Outlines

00:00

🎵 Introduction to AI Music Creation 🎵

The video begins with an introduction to the process of creating music using AI, specifically focusing on generating sung vocals with harmonies without the need for real singers or traditional AI generation tools. The host discusses the controversial nature of AI in music and acknowledges the strong opinions surrounding its use. The video aims to walk viewers through the process of creating a music track using Band-in-a-Box and Ace Studio, starting with a recognizable song called 'So Nice'. The host emphasizes the educational purpose of the demonstration and the intention to respect copyright laws.

05:09

🎶 Style Selection and Melody Conversion 🎶

The host demonstrates how to use Band-in-a-Box to change the musical style of the song 'So Nice' to a Samba style, which fits the groove of the song. The focus is on selecting real tracks for a more realistic sound. The melody track is then exported as a MIDI file, and the music track is exported as a wave file. These files are then imported into Ace Studio, where the melody is converted into sung lyrics. The host guides viewers on how to assign voices to the melody in Ace Studio and adjust the length of the notes and lyrics to fit the song.

10:09

🎤 Vocal Assignment and Lyric Integration 🎤

In Ace Studio, the host assigns a voice to the singer track and begins the process of adding lyrics to the melody. The lyrics for 'So Nice' are obtained from an online search and pasted into the software. The host discusses the importance of aligning the lyrics with the melody and making necessary adjustments to the MIDI file. The process involves some trial and error as the host corrects duplicate syllables and adjusts the length of notes to fit the lyrics properly.

15:11

🎼 Generating Harmony and Vocal Effects 🎼

The host shows how to generate a harmony track in Band-in-a-Box and then convert it into a melody track. This harmony track is then exported as a MIDI file and imported into Ace Studio. In Ace Studio, the host uses the software's features to create vocal doubles, which add depth to the vocals by slightly altering pitch and timing. The host also experiments with changing the octave of the harmony track for a different effect and discusses the process of exporting the tracks for further editing in a digital audio workstation (DAW).

20:11

🎚️ Voice Selection and Mixing in Adobe Audition 🎚️

The host moves on to using Kit's a, a voice conversion tool, to change the lead vocal sound. Various voice models are tested, and the host selects a powerful female voice for the lead. The newly converted vocal track is then exported and imported into Adobe Audition for further mixing and effects processing. The host emphasizes the importance of exporting individual tracks for more control during the mixing process and demonstrates how to create a solo section in Band-in-a-Box to serve as a vocal jazz outro for the song.

25:13

🎉 Final Mixing and Encouragement to Subscribe 🎉

The video concludes with the host finalizing the vocal tracks in Adobe Audition, adding effects like reverb to enhance the vocals. The host reflects on the creative process and encourages viewers to learn how to use these tools together to create unique music. A call to action is made for viewers to subscribe to the channel for more content on the creative uses of AI in various fields, including music, video, art, and animation.

Keywords

AI Music Kitbash

AI Music Kitbash refers to the creative process of using artificial intelligence tools to generate music, specifically in this video, to create a track with sung vocals and harmonies without using real singers or traditional AI generation tools. The term 'kitbash' implies a DIY approach where different elements are combined in a non-traditional way to achieve a unique result. In the video, the creator discusses the use of various AI tools to generate a music track, demonstrating the potential of AI in the music industry.

MIDI

MIDI (Musical Instrument Digital Interface) is a protocol for communicating musical information between digital instruments and computers. It is used extensively in the music production process to create, edit, and transmit music performance data. In the context of the video, MIDI files are used to represent the melody and harmony of the song, which are then processed and converted into sung vocals using AI software.

Harmonies

Harmonies in music are the additional notes or chords that are sung or played along with the main melody to create a fuller and richer sound. They are a fundamental aspect of music that adds depth and texture to a composition. The video focuses on generating vocal harmonies using AI, showcasing how technology can be used to create complex musical arrangements that would traditionally require multiple singers or instruments.

Band in a Box

Band in a Box is a software program designed to generate accompaniments for musical compositions. It allows users to input chords and select musical styles to create a full instrumental backing track. In the video, Band in a Box is used to create the music track for the song 'So Nice,' demonstrating how AI can be used to generate a complete instrumental arrangement from a simple chord progression.

Ace Studio

Ace Studio is an AI-driven music production tool that can convert MIDI melodies into sung vocals. It uses a variety of voice models to synthesize vocals in different languages and styles. In the video, Ace Studio is used to take the MIDI melody generated by Band in a Box and transform it into a vocal track, showcasing the ability of AI to mimic human singing.

Vocal Harmonies

Vocal harmonies are the simultaneous singing of multiple notes or melodies by different voices, creating a layered and often richly textured sound. They are a key element in many musical genres, from pop to classical. The video script discusses the creation of vocal harmonies using AI, which is a significant technological advancement as it allows for the generation of complex vocal arrangements without the need for multiple singers.

Copyright Information

Copyright information refers to the legal rights that creators hold over their work, which includes music, literature, and other forms of creative expression. In the context of the video, the creator mentions the importance of having copyright information on screen when using a pre-existing song like 'So Nice' to demonstrate the capabilities of AI music tools. This is to ensure that the educational use of the song does not infringe on any copyright laws.

Solo

A solo in music is a piece or section of a piece that is performed by a single voice or instrument, often showcasing the skill of the performer. In the video, the creator discusses adding a solo to the end of the song using Band in a Box to generate a unique and interesting conclusion to the track. The solo is an important element in jazz and other musical styles, providing a moment for individual expression within a group performance.

Adobe Audition

Adobe Audition is a digital audio workstation used for recording, editing, and producing audio. It offers a suite of tools for audio professionals and is often used in the final stages of music production to refine and polish tracks. In the video, Adobe Audition is mentioned as the platform where the final mix of the AI-generated music track will be completed, allowing for fine-tuning of the vocals and other audio elements.

Vocal Jazz

Vocal jazz is a genre of jazz music that emphasizes vocal performance alongside the instrumental parts. It often features complex harmonies, improvisation, and a strong sense of rhythm and swing. The video script mentions creating a vocal jazz ending for the song, which involves using AI to generate a unique and stylistically appropriate conclusion to the track that fits within the jazz genre.

Lead Vocal

A lead vocal is the primary vocal part in a song, typically carrying the main melody and lyrics. It is usually the most prominent voice in a musical arrangement. In the context of the video, the lead vocal is generated using AI tools and is later enhanced and mixed with other vocal tracks to create a polished final product. The lead vocal is a critical component in pop and many other music genres, often the voice that listeners focus on.

Highlights

The video demonstrates creating a music track with sung vocals and harmonies using AI, without real singers or instruments.

AI music generation tools are controversial, with strong opinions on their use and quality in the music industry.

The process uses Band-in-a-Box for music accompaniment, allowing the creation of a song from any chord progression.

The video uses a recognizable song, 'So Nice,' for the demonstration, emphasizing it's for educational purposes only.

The song's melody is converted into a MIDI file and then harmonized using Band-in-a-Box's harmony generation feature.

Ace Studio is introduced as a tool for converting the melody line into sung lyrics that can be mixed with the audio track.

Different voice models in Ace Studio are optimized for various languages, with a subset chosen for English lyrics.

Lyrics are manually aligned with the melody in Ace Studio, requiring attention to syllable matching and note length.

The video shows how to create vocal harmonies and doubles in Ace Studio to add depth and richness to the vocal tracks.

Kits is used to convert the AI-generated vocals into a more powerful and emotive lead voice.

Adobe Audition is mentioned as a platform for final mixing and adding effects to the tracks for a polished production.

The process involves exporting individual tracks and compiling them into a multitrack session for detailed editing.

A vocal solo is created in Band-in-a-Box and integrated into the track for a jazzy ending.

The video emphasizes the potential for creative use of AI in music, despite the tools not being perfect.

Different voices and effects are experimented with in Kits to find the best fit for the song's style and mood.

The final mix in Adobe Audition allows for granular control over the vocals, including reverb and volume adjustments.

The video concludes by encouraging viewers to subscribe for more content on the creative uses of AI in various fields.