Chatterbox PT-BR v0.1.0

The first public release of the Chatterbox PT-BR server: Brazilian Portuguese voice cloning through a small FastAPI service.

• Reference upload and chunked speech synthesis returning PCM WAV.
• Readiness based on model state and an explicit unload endpoint.
• Separate model and reference volumes, with a reusable Docker Compose service.
• Quickstart, API contracts and integration guidance for the inference SDK.

Validation includes five CPU contract tests and synthetic reference upload, GPU synthesis and decoded audio inspection. No latency, throughput or speech-quality benchmark is claimed.

Build the image from the tagged source using the quickstart. Model weights are downloaded from the credited upstream checkpoint; this release does not redistribute them.