> ## Documentation Index
> Fetch the complete documentation index at: https://docs.slng.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Supported models

> Full list of TTS and STT models reachable through the SLNG Unified API. Deepgram Nova and Aura, Rime Arcana, ElevenLabs, Cartesia, Sarvam, and more.

Every model on the SLNG platform works through the Unified API. Some run on SLNG infrastructure (SLNG-hosted), others are proxied to the provider's own API.

<Info>
  SLNG-hosted models use the `slng/` prefix in the model identifier. They are optimized for low latency in specific regions and billed through your SLNG account with no separate provider account needed.
</Info>

## Text-to-Speech

<AccordionGroup>
  <Accordion title="Rime Arcana v3" icon="waveform">
    **Provider:** Rime | **Hosting:** SLNG | **Languages:** English, Hindi

    Natural prosody with improved voice quality over v1. French and Spanish support planned.

    | Identifier              | Description |
    | ----------------------- | ----------- |
    | `slng/rime/arcana:3-en` | English     |
    | `slng/rime/arcana:3-hi` | Hindi       |

    Best for: conversational agents, high-quality voice output.
  </Accordion>

  <Accordion title="Rime Arcana v1" icon="waveform">
    **Provider:** Rime | **Hosting:** SLNG | **Languages:** Arabic, English, French, German, Spanish

    150+ voices across 5 languages. Uses unversioned identifiers (`arcana:` instead of `arcana:3-`).

    | Identifier            | Description |
    | --------------------- | ----------- |
    | `slng/rime/arcana:en` | English     |
    | `slng/rime/arcana:es` | Spanish     |
    | `slng/rime/arcana:fr` | French      |
    | `slng/rime/arcana:de` | German      |
    | `slng/rime/arcana:ar` | Arabic      |

    Best for: broad language coverage, high-volume production.
  </Accordion>

  <Accordion title="Deepgram Aura 2" icon="waveform">
    **Provider:** Deepgram | **Hosting:** SLNG + Deepgram | **Languages:** English, Spanish

    Available both SLNG-hosted (low latency) and proxied to Deepgram (more languages). The `model` field is required to pick a voice; there is no default.

    | Identifier                | Description                       |
    | ------------------------- | --------------------------------- |
    | `slng/deepgram/aura:2-en` | SLNG-hosted (English)             |
    | `slng/deepgram/aura:2-es` | SLNG-hosted (Spanish)             |
    | `deepgram/aura:2`         | Proxied to Deepgram (7 languages) |

    Best for: low-latency voice apps, multilingual IVR.
  </Accordion>

  <Accordion title="Cartesia Sonic 3" icon="waveform">
    **Provider:** Cartesia | **Hosting:** Cartesia | **Languages:** 40+

    Streaming TTS with controls for speed, volume, and emotion.

    | Identifier         | Description         |
    | ------------------ | ------------------- |
    | `cartesia/sonic:3` | Proxied to Cartesia |

    Best for: real-time voice agents, expressive multilingual output.
  </Accordion>

  <Accordion title="Murf Falcon" icon="waveform">
    **Provider:** Murf | **Hosting:** Murf | **Languages:** 16 locales

    Real-time multilingual TTS over WebSocket with multiple encodings and sample rates.

    | Identifier            | Description     |
    | --------------------- | --------------- |
    | `murf/murftts:falcon` | Proxied to Murf |

    Best for: low-cost voice agents, IVR, assistants.
  </Accordion>

  <Accordion title="KugelAudio Kugel" icon="waveform">
    **Provider:** KugelAudio | **Hosting:** KugelAudio | **Languages:** 26

    87 voices with expressiveness control across 26 languages. Three variants trade off latency and quality.

    | Identifier                 | Description                    |
    | -------------------------- | ------------------------------ |
    | `kugelaudio/kugel:1-turbo` | Lowest latency                 |
    | `kugelaudio/kugel:1`       | Higher quality, higher latency |
    | `kugelaudio/kugel:2`       | Latest version, EU region      |

    Best for: multilingual voice agents, content creation.
  </Accordion>

  <Accordion title="Soniox TTS v1" icon="waveform">
    **Provider:** Soniox | **Hosting:** Soniox | **Languages:** 50+

    Streaming and one-shot synthesis across 50+ languages.

    | Identifier         | Description       |
    | ------------------ | ----------------- |
    | `soniox/tts-rt:v1` | Proxied to Soniox |

    Best for: multilingual streaming, long-form synthesis.
  </Accordion>

  <Accordion title="Sarvam Bulbul v3" icon="waveform">
    **Provider:** Sarvam AI | **Hosting:** Sarvam AI | **Languages:** 11 Indian languages

    Purpose-built for Indian languages with 30+ speaker voices.

    | Identifier         | Description          |
    | ------------------ | -------------------- |
    | `sarvam/bulbul:v3` | Proxied to Sarvam AI |

    Best for: India-focused voice apps, regional language support.
  </Accordion>
</AccordionGroup>

## Speech-to-Text

<AccordionGroup>
  <Accordion title="Deepgram Nova 3" icon="microphone">
    **Provider:** Deepgram | **Hosting:** SLNG + Deepgram | **Languages:** English, Spanish, Hindi

    Available both SLNG-hosted and proxied.

    | Identifier                   | Description                           |
    | ---------------------------- | ------------------------------------- |
    | `slng/deepgram/nova:3-en`    | SLNG-hosted (English)                 |
    | `slng/deepgram/nova:3-es`    | SLNG-hosted (Spanish)                 |
    | `slng/deepgram/nova:3-hi`    | SLNG-hosted (Hindi)                   |
    | `slng/deepgram/nova:3-multi` | SLNG-hosted (auto-detect)             |
    | `deepgram/nova:3`            | Proxied to Deepgram (17 languages)    |
    | `deepgram/nova:3-medical`    | Proxied to Deepgram (medical English) |

    Best for: low-latency streaming, voice agents.
  </Accordion>

  <Accordion title="Deepgram Nova 2" icon="microphone">
    **Provider:** Deepgram | **Hosting:** Deepgram | **Languages:** 36

    Proven model with VAD, diarization, and smart formatting.

    | Identifier        | Description         |
    | ----------------- | ------------------- |
    | `deepgram/nova:2` | Proxied to Deepgram |

    Best for: production transcription, multi-speaker audio.
  </Accordion>

  <Accordion title="Soniox Speech AI v4" icon="microphone">
    **Provider:** Soniox | **Hosting:** Soniox | **Languages:** 60+

    Real-time STT with speaker diarization, automatic language identification, and configurable endpoint detection. Returns native Soniox token frames.

    | Identifier               | Description       |
    | ------------------------ | ----------------- |
    | `soniox/speech-ai:rt-v4` | Proxied to Soniox |

    Best for: multilingual streaming, multi-speaker conversations.
  </Accordion>

  <Accordion title="Reson8 STT v1" icon="microphone">
    **Provider:** Reson8 | **Hosting:** Reson8 | **Languages:** 9 European

    Real-time STT with word-level timestamps, confidence scores, and partial results. Dutch, English, French, German, Italian, Polish, Portuguese, Spanish, Swedish.

    | Identifier            | Description       |
    | --------------------- | ----------------- |
    | `reson8/reson8stt:v1` | Proxied to Reson8 |

    Best for: European-language voice agents, low-latency transcription.
  </Accordion>

  <Accordion title="Sarvam Saaras v3" icon="microphone">
    **Provider:** Sarvam AI | **Hosting:** Sarvam AI | **Languages:** 24

    Speech recognition for 24 Indian and European languages with code-mixing support. HTTP-only; not available over WebSocket.

    | Identifier         | Description          |
    | ------------------ | -------------------- |
    | `sarvam/saaras:v3` | Proxied to Sarvam AI |

    Best for: India-focused batch transcription, code-switched audio.
  </Accordion>
</AccordionGroup>

## Choosing a model

| Use case                   | Recommended STT           | Recommended TTS         |
| -------------------------- | ------------------------- | ----------------------- |
| Voice agents (low latency) | `slng/deepgram/nova:3-en` | `slng/rime/arcana:3-en` |
| Multilingual transcription | `deepgram/nova:3`         | `cartesia/sonic:3`      |
| Indian languages           | `slng/deepgram/nova:3-hi` | `sarvam/bulbul:v3`      |
| Batch transcription        | `deepgram/nova:3`         | —                       |
| Medical transcription      | `deepgram/nova:3-medical` | —                       |

For voices, regions, and detailed model specs, see the [Model Catalog](/models).