Voice Synthesis and Alteration with AI

“Babylon Shackles” Vocoflex Demonstration 1

“Babylon Shackles” Vocoflex Demonstration 2

Who is this for? Singers, voice teachers, and creators interested in experimenting with synthetic voices and vocal transformation tools.
What will I learn? How to use AI to create, shape, and sing with digital voices—whether for performance, pedagogy, or creative exploration.

Hear Lucy Welsh explain how to use Synth V and demonstrate singing with backup parts she created.

Synthesizer V

Synth V emerged as a central tool for exploring voice modeling, practice, and pedagogy. Students praised Synth V for its expressive control over vibrato, tone, and articulation, allowing them to match or challenge its performance.

You can use Synth V in voice training to:

  • Demonstrate vocal style using Synth V voices

  • Model a “target” sound to emulate or contrast

You can use Synth V as a practice tool to:

  • Create harmony parts for rehearsal and performance

  • Replace missing singers in ensemble works

  • Emulate preferred AI vocal sounds using Synth V

  • Set up different voices to practice interval identification

Maya Zepeda composes with Musescore and AIVA and can hear her vocal lines performed by Synth V singers.

Synthesizer V + AIVA/Suno

Combining Synth V voices with Suno or AIVA instrumental tracks gave students tools to perform and record full songs without access to live collaborators. These setups served as rehearsal companions, artistic partners, and creative outlets. You can try to:

  • Write backing vocals with AI instead of overdubbing

  • Hear compositions in different vocal styles before sending to live singer

  • Get specific tone, emotion, or vocal style you want, which is difficult with generative tools like Suno

Vocoflex Feature Talkthrough

Vocoflex

Vocoflex allowed students to transform their own voices into stylized performances—pitch-shifting, revoicing, and even gender-morphing for creative expression.

  • Used for recorded and real-time transformation

  • Enables singers to hear their voice in different stylistic and emotional contexts

  • Allows creation and demonstration of vocal warmups

  • Demonstrates vocal flexibility for pedagogy and interpretation

  • Listen to AI sing compositions to decide sound/style

  • Use VocoFlex to render compositions with “your” voice