> ## Documentation Index > Fetch the complete documentation index at: https://vida.io/docs/llms.txt > Use this file to discover all available pages before exploring further. # Voice and speech mode > Choose a voice that fits your brand, language, and calling experience. Your Agent's voice affects clarity, pacing, personality, and how customers experience the conversation. ## Choose a compatible voice Select the Agent's language and speech mode first. Then open the voice settings, listen to the available samples, and choose a compatible voice. The available list can vary by account, plan, language, and speech mode. ## Standard and speech-to-speech modes Standard speech mode handles listening and speaking as separate steps. It offers broad voice choices and separate speech-recognition controls. Speech-to-speech modes use one real-time model to understand and produce audio. Vida supports OpenAI and Google Gemini providers, with upcoming OpenAI support for `gpt-live-1`. These modes can feel more conversational, but use a different set of compatible voices and controls. ## Tips for choosing a voice * **Match your brand:** An energetic voice may fit a casual business, while a calm voice may fit professional services. * **Prioritize clarity:** Test phone numbers, addresses, names, and important instructions. * **Consider your audience:** Choose a pace and tone that callers can understand comfortably. * **Compare real calls:** A short sample does not reveal interruptions, longer responses, or noisy environments. Save the change, test a staged call, and publish only after the voice works well with the selected language and instructions. Selecting voices programmatically? See [Language, voice, and speech](/docs/api-reference/agent-guides/language-voice-and-speech).