Voice AiLunaPro supports Deepgram, OpenAI, Google, Azure Speech, AssemblyAI, Speechmatics, Cartesia, Gladia, Soniox, Sarvam, Smallest AI, Hugging Face, ElevenLabs, and AiLunaPro transcribers. You can take a look at the providers documentation of which language to select for your language requirements.
Transcriber providers receive the data needed to transcribe speech, such as call audio, language settings, keyterms, and request metadata. Review the provider’s data processing, retention, model training, and regional hosting policies before using sensitive data.
Deepgram Flux language hints
With the flux-general-multi model, Language hints biases transcription toward the languages you expect callers to speak. Pick several when callers switch between known languages, for example Hindi and English. Hints are combined with the Language setting, and leaving both on auto-detect lets Flux detect the language on its own.
Hints only help within the languages Flux multilingual supports. They do not add new ones. See Deepgram’s language prompting guide.
In AiLunaPro mode, language hints are also available when you select auto-detect (multi) or a Flux-supported language. For bilingual English/Spanish calls, choose auto-detect and check English and Spanish under Language hints. These hints bias recognition toward both languages; they do not guarantee that other languages will be excluded.
Selecting a specific Flux-supported language combines that language with your hints. Hints are hidden and unused when you select a language that AiLunaPro routes to its standard transcriber. 