Voice & Audio Settings

Voice and audio settings determine how your agent sounds, how it understands speech, and which language model powers its reasoning. Getting these right is essential for creating a natural, responsive conversational experience.

Text-to-Speech (TTS)

The TTS provider converts your agent's text responses into spoken audio. Cloudax Connect supports multiple providers, each offering distinct voice characteristics and performance profiles.

Voice Library

The voice library includes hundreds of pre-made voices across different accents, ages, and styles. Browse and preview voices from the Personalities page, then select one in your agent's settings. You can also use a custom voice ID if you have created a custom voice.

Voice Parameters

Depending on the provider, you can adjust the following parameters to fine-tune how your agent sounds:

• Stability: controls how consistent the voice sounds across utterances. Higher values produce more uniform delivery; lower values introduce natural variation.
• Similarity boost: how closely the output matches the original voice sample. Higher values sound more like the reference voice but may reduce expressiveness.
• Style: adjusts the emotional expressiveness of the voice. A higher style value makes the agent sound more animated and engaging.
• Speed: controls the speaking rate. Values above 1.0 speed up delivery; values below 1.0 slow it down.

TTS Model

The voice runs on one of several ElevenLabs model tiers, which trade latency against audio quality and language coverage. These are the options in the model dropdown in your agent's voice settings:

• Expressive: the default for new agents. The newest and most expressive real-time model, with ultra-low latency, a natural emotional range and support for more than 90 languages.
• Flash 3: the most expressive real-time model, with context-aware delivery and support for more than 70 languages. Choose it when natural emotion and conversational nuance matter most.
• Flash 2.5: fast and efficient voice generation with the lowest latency, and supports 32 languages. A fast option for snappy, high-volume agents.
• Flash 2: the previous-generation Flash model. It is English-only, but tends to produce fewer voice artefacts, so it can be a strong choice for English agents where audio cleanliness matters more than language coverage.
• Turbo: higher quality and only slightly slower than Flash, with support for 32 languages. It delivers better audio fidelity and reduces artefacts such as glitches, clipped words, robotic warbling, and mispronunciations.

Speech-to-Text (STT)

The STT provider transcribes the caller's speech into text so your agent can understand and respond. Accurate transcription is critical for conversation quality.

STT Models

Choose the model that best matches the language(s) your callers will speak:

• Conversational : supports 10 languages (English, Spanish, French, German, Hindi, Russian, Portuguese, Japanese, Italian, and Dutch). Pick a single language for best accuracy, multiple languages if your callers switch between them in the same call, or none to auto-detect. Use this when you want Conversational mode.
• Other languages (default): supports 50+ locales, including Multilingual auto-detect across about 55 languages with code-switching. This is the default for new agents. See Supported languages for the full list.
• Scribe: supports 90+ languages with high-accuracy multilingual transcription. See Supported languages for the full list.

Supported languages

Speech recognition language coverage depends on the mode you pick in the agent’s voice settings. Conversational covers 10 languages, Other languages covers Multilingual auto-detect across about 55 languages plus 50+ single-language locales, and Scribe covers 90+. Full lists for each mode are below.

• Conversational: English, Spanish, French, German, Hindi, Russian, Portuguese, Japanese, Italian, Dutch. Pick one language for best accuracy, several if callers switch mid-call, or Multilingual / empty to auto-detect across these ten.
• Other languages: Multilingual auto-detect across about 55 languages (with code-switching), plus single-language locales: English (US), English (UK), English (Australia), English (India), English (New Zealand), Arabic (Saudi Arabia), Arabic (UAE), Arabic (Egypt), Arabic (Morocco), Arabic (Qatar), Arabic (Kuwait), Arabic (Jordan), Arabic (Lebanon), Arabic (Iraq), Arabic (Algeria), Arabic (Tunisia), Belarusian, Bulgarian, Bosnian, Catalan, Croatian, Czech, Danish, Dutch (Netherlands), Dutch (Belgium / Flemish), Estonian, Finnish, French (France), French (Canada), German (Germany), German (Switzerland), Greek, Hungarian, Italian, Latvian, Lithuanian, Macedonian, Norwegian, Polish, Portuguese (Brazil), Portuguese (Portugal), Romanian, Russian, Serbian, Slovak, Slovenian, Spanish (Spain), Spanish (Latin America), Swedish, Turkish, Ukrainian, Bengali, Chinese (Cantonese), Chinese (Mandarin, Simplified), Chinese (Mandarin, Traditional), Hindi, Indonesian, Japanese, Kannada, Korean, Malay, Marathi, Tamil, Telugu, Tagalog (Filipino), Vietnamese.
• Scribe: Afrikaans, Amharic, Arabic, Armenian, Assamese, Asturian, Azerbaijani, Belarusian, Bengali, Bosnian, Bulgarian, Burmese, Cantonese, Catalan, Central Kurdish, Chichewa, Chinese, Croatian, Czech, Danish, Dutch, English, Estonian, Filipino, Finnish, French, Fulah, Galician, Ganda, Georgian, German, Greek, Gujarati, Hausa, Hebrew, Hindi, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kabuverdianu, Kannada, Kazakh, Khmer, Kyrgyz, Korean, Lao, Latvian, Lingala, Lithuanian, Luo, Luxembourgish, Macedonian, Malay, Malayalam, Maltese, Maori, Marathi, Mongolian, Nepali, Northern Sotho (Pedi), Norwegian, Occitan, Oriya, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Serbian, Shona, Sindhi, Slovak, Slovenian, Somali, Spanish, Swahili, Swedish, Tajik, Tamil, Telugu, Thai, Turkish, Ukrainian, Umbundu, Urdu, Uzbek, Vietnamese, Welsh, Wolof, Xhosa, Zulu.

STT Settings

• Language: set the expected language of the caller. This improves transcription accuracy, especially for non-English calls.
• Keyterms: provide custom vocabulary (product names, brand terms, industry jargon) to help the STT engine recognise words it might otherwise miss or misspell.
• Endpointing delay: the amount of silence (in milliseconds) before the STT engine considers the caller's turn complete. A shorter delay makes the agent more responsive; a longer delay reduces the chance of the agent interrupting the caller mid-thought.

Language Models

The language model (LLM) is the brain of your agent: it interprets the caller's intent, follows your system prompt, and generates responses. Available models, listed roughly from newest to oldest:

• 6.1 Terra (default): recommended for new agents. No reasoning effort. Strong visual understanding and multi-step work, with a large context window. Prefer this for new work and when migrating off a retiring model.
• 6 Luna: faster, lower-cost no-reasoning tier when you do not need 6.1 Terra’s extra capability. Large context window; suited to high-volume live conversations.
• 5.3 Chat: a fast conversational model from the 5-series. Strong general capability with low latency. Retired 10 August 2026. Migrate to 6.1 Terra.
• 4.1: the full-size 4.1 model. Use when the agent needs maximum instruction-following accuracy and you can afford a little more latency and cost than Mini. Retires 14 January 2027. Migrate to 6.1 Terra.
• 4.1 Mini: previous default. Still a solid balance of speed and intelligence for existing agents. Retires 14 January 2027. Migrate to 6.1 Terra.
• 4.1 Nano: the fastest 4.1 variant for simple, high-volume interactions. Retires 23 October 2026. Migrate to 6.1 Terra as soon as practical.
• 4o: multimodal model with strong audio-aware capabilities for agents that need particularly natural turn-taking. Retires 14 January 2027. Migrate to 6.1 Terra.
• 4o Mini: multimodal capabilities at a compact size for agents that process both audio and text on a tighter latency budget. Retires 14 January 2027. Migrate to 6.1 Terra.

6.1 Terra is the recommended starting point for almost every agent. Switch to 6 Luna when you need lower cost or the lowest latency and live tests show Terra’s extra capability is not required. Keep a retiring model only while you validate behaviour after a prompt or tool change, then switch before its retire date.

Model retire dates

These dates are when the model stops being available for new and existing agents. After the date, switch the agent to the replacement model and redeploy.

Temperature

The temperature setting controls how creative or deterministic your agent's responses are. A value of 0 produces highly consistent, predictable responses. A value of 1 introduces more variation and creativity. For most telephony use cases, we recommend a temperature of 0.2 to keep responses focused and on-script.