BRAINIALL Speech provider RC: pt-BR diarization and TTS

[fasuizu-br]

BRAINIALL has published a release candidate for a Dify model provider focused on Brazilian
Portuguese speech workflows:

• speech-to-text fixed to pt-BR with speaker diarization;
• text-to-speech with three allowlisted pt-BR voices;
• fixed HTTPS origin, no redirects, zero automatic retries, explicit timeouts and sanitized
  provider errors.

Repository: https://github.com/fasuizu-br/brainiall-dify-provider

Release candidate: https://github.com/fasuizu-br/brainiall-dify-provider/releases/tag/v0.1.0-rc.1

The package passed 41 mocked transport/SDK tests and Dify CLI packaging validation. The
valid-key live smoke test inside Dify is still pending, so this is explicitly an evaluation
candidate rather than a production certification or Marketplace listing.

The plugin is MIT-licensed and free. BRAINIALL API usage is metered and may be charged after
included credits; audio and synthesis text are sent to the BRAINIALL API under the linked
privacy notice.

Useful beta feedback would be reproducible results for: installation, pt-BR diarized audio,
WAV synthesis, and Dify version compatibility. Please report issues in the source repository
without including API keys or user content.