French voicemode model

### Describe the feature or problem you'd like to solve

Being able to use a french speech-to-text model through /voice

### Proposed solution

_No response_

### Example prompts or workflows

_No response_

### Additional context

_No response_

Start at the /voice entry point and trace how the speech-to-text model is selected. Determine how French model support should be exposed and verify that a French voice workflow completes successfully. The issue provides no named files, tests, or proposed implementation.

索引モデルが issue の本文から書いたものです。