### Answer Retell doesn't expose full control over the STT model selection through their API [(1) ]() . The system uses their hosted model by default, and you're limited to the default configuration unless you implement a more custom setup . However, you can configure some STT-related parameters through the agent settings [(2) ]() [(3) ]() : - **language**: Specify the speech recognition language (e.g.,

en-US

,

en-GB

,

multi

for multilingual) - **boosted_keywords**: Provide a customized list of keywords to bias the transcriber model, making specific words more likely to get transcribed correctly [(2) ]() [(3) ]() [(4) ]() - **stt_mode**: Set to "accurate" for optimized transcription accuracy [(1) ]() - **custom_stt_config**: Configure provider and endpointing settings For improving transcription quality beyond these settings, you can adjust denoising mode and enable transcription mode optimized for accuracy [(5) ]() .

​ To learn more, see the sources I considered: [[1] Discord message from @sloanluvv on 2026-04-16](https://discord.com/channels/1186424376300679209/1186462578310910112/1494346885652152351) [[2] Create Voice Agent](https://docs.retellai.com/api-references/create-agent) [[3] Update Voice Agent](https://docs.retellai.com/api-references/update-agent) [[4] Step 2: Configure the basic settings](https://docs.retellai.com/build/single-multi-prompt/configure-basic-settings) [[5] Increase transcription accuracy](https://docs.retellai.com/reliability/wrong-transcript) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.