I work as a fractional product marketer and growth operator for subscription software companies, partnering directly with founders to clarify product positioning, improve conversion across the customer lifecycle, and make revenue performance legible.

Earlier in my career, I led product marketing at a B2B scheduling SaaS where I facilitated a 20% increase in trial-to-paid conversion and ~$700K in incremental ARR over three years. 

Today, I work with a small number of companies at a time, embedding closely with product and engineering to improve clarity, conversion, and retention, especially in situations where growth has plateaued or metrics are difficult to trust.

Former clients include Suno, WAVS, Kits.AI, Neural Frames, Lemonaide, Sesh.fm, NeuTune, Water & Music, ACE Studio, Audio Design Desk, RipX, HookTheory, Samplab, and others.

Ezra Sandzer-Bell

Here’s the New Years Guide to AI Music Generators in Q1 2026, with top companies and reporters to watch.🌟

🦍 Suno: The 800 pound gorilla

$200M ARR in Q4 2025, $2.45B valuation.

Suno scraped volumes of music without a license, claimed fair use in court, and signed a deal with WMG at the end of 2025.

Meanwhile, they've gained first mover advantage and attracted world class talent. SUNO Studio (Suno + WavTool) is the best AI DAW on the market, if we're talking about music quality and software UX.

Suno allows commercial use and grants song ownership to paid users, despite a moment of confusion in December.

 💐 Runners Up: Alternatives to Suno

↝ Udio: High quality AI song generator, trained on unlicensed music just like Suno.

Their AI voice models improved last year, but a new UMG partnership dictates a no download policy that upset their base.

If musicians can't use Udio for stems, and hustlers can't turn AI songs into royalty checks from DSPs, who is Udio selling to?

They might pivot to AI streaming, like KLAY.

↝ ElevenLabs: Single track audio timeline editor. Generates sfx/foley and music from a single prompt. Best for media teams on a limited budget.

↝ Mozart AI: AI DAW and music model hub. Acts as a wrapper for the ElevenLabs model, exposing a multitrack editor and instrument stems. 

↝ Aiode: AI DAW in public beta. Contextual stem generation, static instrument models (think Lemonaide Music but audio instead of MIDI).

↝ Producer.ai (formerly Riffusion): AI chat interface with text to music generation. Songs are generic, UX is clean.

↝ Controlla: AI singing voice company with song generation tool and B2B API.

↝ Moises Studio: Honorable mention. AI DAW with bland, generic audio but decent market visibility due to parent company.

🥷 Stealth companies to watch for in 2026

↝ KLAY: AI streaming service partnering with majors, not launched yet.

↝ MixAudio by Neutune: AI music gen company, specializing in IP attribution for genAI models.

↝ Stability AI's Stable Audio: Open source sample generator, new UMG partnership.

↝ Splice: New partnership with UMG announced at end of year. No details yet about the model or how it might fit into their product ecosystem.

↝ OwlDuet: Voice operated DAW, waitlist only but I was told they’re active.

🙋🏻‍♂️ LinkedIn reporters I follow for AI music industry updates:

1. Christopher Wieduwilt (The AI Musicpreneur) - Weekly news bites
2. Cherie Hu - Founder at Water & Music, deep AI music industry data
3. Kristin Robinson - Investigative journalist with AI music focus at Billboard
4. Virginie Berger - AI music IP attribution expert, Forbes journalist
5. Jordi Pons - AI music research scientist and artist, Stable Audio
6. Rithik Kundu - Founder and event coordinator for NYU's AI music program
7. Mark Redito & John von Seggern - AI music focused podcast hosts
8. Drew Thurlow - Music and AI professor at Berklee, ex Sony/Warner

💫 Follow me here for ongoing AI music news and industry insights.

Music software founder turned fractional PMM. Growth advisor, product manager and copywriter for music tech startups.

Dreamtonics just launched a powerful, ethically-trained AI Choir tool for composers and music producers. 🚀

The new AI choir feature expands on their core standalone app, Synth V Studio Pro 2. 

I'll explain how this all fits into the greater music tech ecosystem below.👇 

🎙️ Dreamtonics: Synth V Pro 2 for musicians

Dreamtonics published the original Synth V software back in 2018. It took them six years and lots of R&D to arrive at version two.

Synth V Pro 2 came out in March 2025, focused on solo voice synthesis. It's a serious musician tool, with a MIDI piano roll and phoneme-level vocal timbre control.

Polyphonic arrangements were *technically* possible in the 2025 version. But stacked, solo vocal stems didn’t deliver the vocal realism of a live choir.

Dreamtonics spent two years researching technical solutions, recording real human choirs (consensually licensed) to train their AI models.

On January 15th, they released three choir voice collections with support for up to 16 vocalists singing in English, Spanish, Japanese, Korean, Mandarin, or Cantonese.

🎹 Better than Suno: Control every note and vocal articulation

Dreamtonics's Synth V Pro 2 outperforms Suno Studio in two major ways; note and timbre control.

Suno does not provide stem-level note manipulation (i.e. autotone). They have a newer audio-to-MIDI feature, but those MIDI notes are divorced from the timbre and lyrics of the original audio file.

Other AI audio-to-midi tool like Hit'n'Mix - RipX or Samplab, or even Celemony Software GmbH's Melodyne Autotune software, let you modify notes in the original audio stem. Suno has not rolled this out yet.

Dreamtonics supports note dictation for custom lyrics AND allows users to edit the pronunciation of every syllable. None of the above audio-to-midi tools support vocal timbre control of this kind.

🗺️ Synth Voices: Mapping the territory

Old school MIDI synth choir patches were limited to vowel tones. Those cheeky "male choir ooh" sounds worked well enough for classic JRPGs and 80s synth ballads, but they're outdated and rarely used anymore.

Modern music software has improved on the timbre of the synth vocals, but are still limited to vowel sounds or random single-word triggers. 

Industry leaders like Spitfire Audio and ARTURIA don't support lyric inputs or phoneme control. They’re ordinary synths.

Choir: Omnia by Native Instruments supports syllable-by-syllable vocal sampling without an AI model, with far less control over notes and pronunciation. 

Choir Omnia costs $400, making it more expensive than Synth V Studio Pro 2 ($99) and a $150 Dreamtonics choir voice collection. 

🥊 TIMEDOMAIN's ACE Studio 2 is the only other AI singing voice DAW that competes with Dreamtonics today. They have a piano roll and AI choir feature.

Controlla and Kits.AI are two simpler, popular AI choir tools for web browser. Controlla is less expensive than Kits and grants users more output minutes per subscription tier.

Aurora Flight Sciences

Integrated Product Team lead for LIDAR-based Landing Hazard Avoidance (LHA) for the Wisk Gen6 aircraft. Lead software engineer for a globally distributed team developing Detect and Avoid (DAA) analysis and simulation tools, reporting to the Boeing CTO.

Senior Autonomy Engineer

Anduril Industries

Lead software engineer for integration of the WISP/SkyFence sensor into Lattice. Developed a multi-target tracker regression testbed.

Estimation and Tracking Engineer

Scientific Systems

R&D areas: Machine learning for visual tracking and system identification; multi-agent route planning; image-aided navigation for GPS-denied environments; 3D scene reconstruction using one or more cameras; helicopter landing zone identification; graph-based SLAM; human motion modeling for personal navigation and tracking applications; novel techniques for extreme image compression; sensor fusion; aircraft state distribution prediction for DAA; unexploded ordinance identification using ground-penetrating RADAR; Cruise Missile Auto Routing System (CMARS); FORESTER Target Density Tracker (TarDenT). David was the principal investigator for the Trajectory Optimization Manager for Multiple Algorithms and Sensors (TOMMAS). He also developed middleware that extends the capabilities of the Robotics Operating System (ROS) and the Mission-Oriented Operating Suite (MOOS). Funded by DARPA, AFRL, ARO, NAVAIR, NASA.

Senior Research Engineer

MIT CSAIL

Real-time tracking of natural visual features. Mentored a senior design team working on the Modular Rover for Extreme Terrain Access (MoRETA). Funded by NASA.

Student Researcher

Draper Laboratory

Computer vision for pose estimation and active navigation. Contributions to the Rubble Crawler, High Mobility Tactical Micro Robot, and Personal Navigation System. Implemented image stabilization for embedded processors. Algorithms for combining inertial and video data. Funded by AFRL.

Research Fellow

UCF Computer Vision Laboratory

Computer vision research, segmentation and tracking, systems for AUVSI competitions. Funded by NSF-REU.

Lockheed Martin

Analysis of flexible body dynamics for flight control. Funded by IR&D.

Internship

NOT SEEKING WORK AT THIS TIME

I can help your company to advance the state-of-the-art of autonomy by leveraging substantial experience in the following technology areas:

* Computer Vision: Machine Learning, Image Registration, Segmentation, Pattern Recognition, Contour Tracking, 6-DoF Tracking
* Sensing & Navigation: Hemispherical FoV Cameras, Gimbal Stabilization, Consumer & Tactical INS/GPS Sensors, Vision-Aided Inertial Navigation, WGS84 Models, Personal Navigation
* Prediction & Control: Reinforcement Learning, Linear & Nonlinear Systems, EKF, UKF, SLAM, Particle Filters
* Planning: Route Planning, Dynamic Task Optimization, Advanced Collision Avoidance, Swarm Planning
* System Design & Engineering: Deep Dive, Agile, Design for Manufacturing, Reliability, Sustainability, Scalability, and Trust-in-Autonomy

Programing Languages and Libraries: C/C++, MATLAB/Simulink, Python, PyTorch, TensorFlow/Keras, OpenCV, OpenGL.

Mechanical Design and Manufacturing: Requirements engineering, statistical process control, material selection, manual and CNC machining, laser cutting, injection molding, thermoforming, casting, water jet, Solid-Works, Pro-Engineer, I-DEAS, AutoCAD, DADS/flex.

taplio

The Battle

Ezra Sandzer-Bell

David Diel

Followers

Posts

Engagements

Likes

Performance ScoreEzra Sandzer-Bell's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Ezra Sandzer-Bell

David Diel

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Ezra Sandzer-Bell's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreEzra Sandzer-Bell's LinkedIn Performance