A songwriter has an awkward question about the last word of a chorus. Should the harmony reach the vowel before the lead singer or hang behind it? Rebooking a session for each possibility would be a poor way to find out. Dreamtonics builds software that lets the writer enter the notes and lyrics, hear a virtual singer perform them, and alter how individual syllables land. That makes its AI especially interesting to musicians who already know what they want to write.
- 01The instrument Synthesizer V Studio 2 Pro connects musical notation and lyrics to a licensed singing voice.
- 02The purchase The editor, additional voices and permission to use a particular voice are separate decisions.
- 03The listening test Carry an exposed harmony line through a real change of key, lyric or phrasing before building a production around it.
01 / ProductA singing instrument with three editable ingredients
A session starts with a score, a voice database and instructions about delivery. The score determines which pitches the singer is being asked to reach and when. The voice gives those notes a recognizable character. Performance controls change details such as timing and timbre without requiring a different melody. The Studio 2 Pro overview describes a neural engine that runs locally, with both standalone operation and integration into a digital audio workstation.
That division is useful on a writing deadline. If a composer decides that the top line should descend at the end of a phrase, the arrangement can change while the lyric stays in place. If the notes are right but the ending sounds abrupt, an editor can work on the pronunciation or timing of that ending. A new render does not require commissioning a new composition. The musician still has to decide which change solves the musical problem.
The Dreamtonics product family also includes Instrument X for writing instrumental parts and Vocoflex for transforming vocal sound. A request to change a recorded performance and a request to write new sung notes are different briefs. Ask the producer whether a recorded take already exists and which aspects must survive the edit. That question usually narrows the software choice before anyone begins comparing the number of voices in a catalogue.
02 / AudienceWhere a virtual session singer earns its place
A songwriter preparing a guide track for a collaborator needs intelligible phrasing and a clear tune. A backing vocalist recording the part later might want cues about entries, breathing and held consonants. The guide can communicate those choices much better than a piano doubling the melody. A final vocal used in a released recording raises additional questions about the sound of the exposed voice, which database was used and which licence covers the release.
A music director arranging a choir has a different test. A harmony that sounds convincing in isolation can obscure the lead when several parts pronounce the same consonant on the same beat. Give an arranger a phrase with a sustained vowel, a moving inner voice and a word ending shared by several singers. Listen to where intelligibility disappears when the voices enter together. A useful editing tool should let the arranger find the offending part quickly enough to keep working musically.
The voice catalogue names the publisher alongside individual voices. Felicia is a Dreamtonics voice, for example, while NYL is published by Eclipsed Sounds. A favourite demonstration therefore identifies a specific instrument to investigate. It cannot tell a composer how another voice will handle a different range, pronunciation or genre. Keep a shortlist based on the register and words in the actual score.
03 / WorkflowWrite and revise the middle harmony before arranging the full chorus
For a proposed studio evaluation, choose four bars of original music that the team is free to use. Include a held syllable, a quick succession of short words and one place where the lead and harmony move in different directions. Put a plain keyboard and a rough lead against the same tempo grid. Enter the intended pitches and lyric for the middle voice, and listen to its first pass against the lead at an ordinary mixing level.
Make the first change small enough to hear. Move a problematic syllable earlier, then listen again from the previous bar. A consonant that appears early on its own can still place the accented vowel exactly on the beat. Changing the whole note to fix that consonant could damage the groove. The phoneme timing controls provide a way to test that distinction. Keep the take that makes the word clear in the mix even if another sounds more striking in isolation.
Next use the retake controls to explore alternate timing or timbre on one phrase. Hold the pitches and the accompaniment still. Save a reference version before choosing a variation so that a later lyric revision does not depend on remembering which setting sounded right. Then transpose the phrase to the key requested by the singer. Moving pitches can place the synthetic voice in a different part of its range, making an otherwise successful edit less useful.
Finally ask a musician who has not seen the piano roll to sing the harmony back. If the reference causes repeated mistakes on the same entry, revisit the note length or diction. For an evaluation using the demo, listen inside the editor. The trial comparison marks file export and commercial voice licensing as full-version capabilities in its comparison table. Reserve the exported stem and release checks for a licensed configuration.
04 / PricingBudget for the editor, then check the specific voice
| Purchase | Displayed price | Scope |
|---|---|---|
| Studio 2 Pro demo | Editor free for 14 days | Individual voice trials last seven days. Plan the listening session accordingly. |
| Studio 2 Pro | $99 once | Full individual editor with a voice choice shown. |
| Additional professional voice | From $79 | Price and rights depend on the selected database. |
Selected English-site dollar prices and licence basis, consulted 19 November 2026. Studio 2 Pro pricing and trial durations.
A perpetual licence changes which costs deserve attention. Rendering another take does not appear as a per-minute charge on the editor price list. A new performer, an additional editor at a second workstation or a plan to deliver isolated material to a sample library can change the cost or eligibility. Start with the smallest combination that covers the people producing the track and the material they will actually deliver.
Dreamtonics’ voice terms cover databases it produces and direct users of partner voices to different terms. The summary permits commercial distribution of musical works made with its voices, describes a single-person licence, and excludes making sample packs or using generated vocals to train machine-learning systems. A release on a record and a pack of reusable vocal samples can contain similar sounds while having different permissions. Check the intended deliverable with the publisher when it falls near that boundary.
05 / DistinctionsWhat changes when the producer owns the score
In Suno’s songwriting workflow, a creator can explore generated songs and compare musical directions. Dreamtonics fits a producer who wants to specify the middle harmony before hearing the result. A useful comparison is to change just one late consonant while keeping the other voices and accompaniment stable. The time and precision required for that particular revision matter more than which first audition produces the most impressive chorus.
Moises’ approach to working with recordings helps frame a different problem. Someone preparing to practise against an existing recording starts with audio containing parts they want to hear differently. A composer building a missing vocal line starts with an intended arrangement. Identify what material already exists, who controls it and whether the desired result is a practice mix, a performance guide or a new part for release.
06 / QuestionsWhich project details must survive the handoff?
A bounced vocal is convenient for the mix engineer. It cannot tell the next arranger which voice, software version and expression edits created the take. Deliver a short note with the session tempo, chosen voice and the exact bar where each exported part begins. If a client expects fully editable files, agree in advance which collaborators need their own licensed tools to make revisions.
Offline rendering in the full editor is attractive to a studio working on unreleased music. Installation, activation and moving a project to a replacement machine are separate operational questions. Rehearse those steps before a booked session and ask the supplier about a studio with multiple editors. Dreamtonics lists corporate multiseat and SDK enquiries separately. Those routes matter when a software developer wants to embed an engine or a facility wants several people to edit the same production.
07 / DecisionChoose the kind of vocal the project needs
A songwriter preparing a guide
Write the hardest few bars first, including the consonants that carry the groove. Ask the intended singer to interpret the guide before investing in a full arrangement.
A producer planning a released harmony
Check the publisher of every database, the number of people editing and the intended musical deliverable. Make a licensed export and hear it beside the lead.
A studio building a reusable voice library
Describe exactly what users would receive and whether they could reuse isolated sounds. Take that description to the publisher before recording the catalogue.
A business worth understanding.
Suggest your business or one you find interesting. Tell us what you want to understand about its product, positioning, design or workflows.
Suggestions are free. Selection and publication stay with the desk.
- Dreamtonics instruments and vocal toolsConsulted
- Synthesizer V Studio 2 Pro features and pricesConsulted
- Voice catalogue and publishersConsulted
- Dreamtonics voice database usage termsConsulted
- Trial and bundle comparisonConsulted
- Dreamtonics company and licensing enquiriesConsulted
