The NVidia documentation shows that the [Nemotron ASR models support `word boosting`](https://docs.nvidia.com/nim/speech/26.05.0/asr/customization/customization.html#word-boosting) to correctly recognize jargon words. Does Foundry Local support that, and if so, how can we implement that? If not, please see this issue as a feature request.
The NVidia documentation shows that the Nemotron ASR models support
word boostingto correctly recognize jargon words.Does Foundry Local support that, and if so, how can we implement that?
If not, please see this issue as a feature request.