Clinical speech · Research framework

More than words.
More than one
kind of privacy.

Transform voice identity and spoken identifiers separately. Keep the conversation in view. Measure what each intervention changes.

Build with AegisVox ↗Watch the demonstration

Python framework · Apache-2.0 · Model-adapter architecture

Prosody. Beyond a transcript.Pauses. Part of the interaction.Roles. Who said what.Context. Worth measuring.
The architecture

Factorize the change.
Follow the evidence.

Changing words and changing a voice solve different problems. The implementation keeps those operations, their artifacts, and their measurements distinct.

01 / IDENTITY

A consistent voice per speaker

Profile speakers, retrieve distinct voice references, and convert all turns with SeedVC. The assignment follows the speaker, not the sentence.

All speaker turns
02 / LEXICAL

Edit the complete affected turn

Plan document-consistent replacements. Render the revised turn with its converted voice reference. Rejoin it with the other converted turns.

Selected complete turns
03 / EVIDENCE

Do not hide a tradeoff in one score

Inspect speaker linkage, replacement realization, retained identifiers, and clinical preservation at the stage where each was measured.

Separate endpoints
78 seconds inside AegisVox

See the conversation move.

The technical film shows consistent speaker mappings, selective complete-turn edits, and the checkpoints between transformation and evaluation.

Illustrative signals, not patient speech or model internals. Silent video. Read the equivalent architecture description.

A real starting point

Try the mechanics.
Bring your models.

A lightweight local example verifies reconstruction without a GPU, API key, or model download.

The demo is synthetic tones, not anonymized speech. Real inference requires authorized models, reference voices, and data.
git clone https://github.com/spraphul/aegisvox.git
cd aegisvox
python3 -m venv .venv
source .venv/bin/activate
pip install -e '.[dev,evaluation]'
aegisvox demo --output runs/first-demo
pytest -q
Installation and first run →
Developer documentation

From inputs to inspectable artifacts.

System architecture ↗Complete-turn contract ↗Models and resident workers ↗Deployment and service API ↗Stage-resolved evaluation ↗Release scope and provenance ↗
A research framework, not a privacy or clinical-safety certificate. Residual linkage, missed identifiers, imperfect edits, and model-based measurement remain important limits.