### Answer Based on the available sources, Retell AI does not currently support using your own custom Speech-to-Text (ASR) model. [(1) ]() In a Discord discussion, when a user asked about specifying their own transcription models, a Retell team member confirmed that Retell uses Deepgram for ASR and noted that "streaming deepgram has less performance even if is the same model." For multilingual or code-switching use cases, a Retell co-founder mentioned they "have been evaluating" true multilingual ASR options like Whisper, but this is not currently available as a configurable option. Check out these potentially useful sources for more details on speech processing: - [Speech Processing]() - [Terminologies]() - [Normalize text for speech]() Would you like me to try to answer generally based on my knowledge, or do you have a specific use case you're trying to address?

​ To learn more, see the sources I considered: [[1] Multilingual or Bilingual Transcription](https://discord.com/channels/1186424376300679209/1230269781161279508) ​

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