> ## 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.

# Text-to-Speech Models

> Browse every TTS model on SLNG: Cartesia, Deepgram Aura, ElevenLabs, KugelAudio, Murf, Rime Arcana, Sarvam, and Soniox voices.

export const ModelCard = ({name, type, provider, hosting, description, features = [], regions = [], href, bestFor, compact = false}) => {
  const PROVIDER_ICONS = {
    Deepgram: "/images/providers/deepgram.svg",
    "Sarvam AI": "/images/providers/sarvam.svg",
    OpenAI: "/images/providers/openai.svg",
    KugelAudio: "/images/providers/kugelaudio.jpeg",
    Rime: "/images/providers/rime.png",
    Soniox: "/images/providers/soniox.jpeg",
    Cartesia: "/images/providers/cartesia.jpeg",
    Murf: "/images/providers/murf.svg",
    Reson8: "/images/providers/reson8.svg"
  };
  const PROVIDER_INITIALS = {
    Deepgram: "DG",
    Soniox: "SX",
    "Sarvam AI": "SA",
    OpenAI: "OA",
    Rime: "RI",
    CanopyLabs: "CL"
  };
  const isSlng = hosting === "slng";
  const typeBadgeClass = type === "stt" ? "model-card__badge--stt" : "model-card__badge--tts";
  const typeLabel = type.toUpperCase();
  const hostingLabel = isSlng ? "SLNG-hosted" : `${hosting}-hosted`;
  const providerIcon = !isSlng ? PROVIDER_ICONS[provider] : null;
  const initials = isSlng ? "S//" : PROVIDER_INITIALS[provider] || provider.slice(0, 2).toUpperCase();
  const subtitle = isSlng ? `By ${provider} via SLNG` : `By ${provider}`;
  const logoContent = providerIcon ? <img src={providerIcon} alt={provider} className="model-card__logo-img" /> : initials;
  const groupRegions = rs => {
    const hasMore = rs.includes("+more");
    const zones = [...new Set(rs.filter(r => r !== "+more").map(r => r.split("-")[0].toUpperCase()))];
    const label = zones.join(" · ");
    return hasMore ? `${label} + more` : label;
  };
  const displayRegions = regions.length > 0 ? groupRegions(regions) : "";
  const cardContent = compact ? <div className="model-card compact">
      <div className="model-card__header">
        <div className={`model-card__logo ${isSlng ? "model-card__logo--slng" : "model-card__logo--provider"}`}>
          {logoContent}
        </div>
        <div className="model-card__header-text">
          <div className="model-card__title-row">
            <span className="model-card__name">{name}</span>
            <span className={`model-card__badge ${typeBadgeClass}`}>{typeLabel}</span>
          </div>
          <div className="model-card__subtitle">{subtitle}</div>
        </div>
      </div>
      {features.length > 0 && <div className="model-card__pills">
          {features.map((f, i) => <span key={i} className="model-card__pill">{f}</span>)}
        </div>}
    </div> : <div className="model-card">
      <div className="model-card__header">
        <div className={`model-card__logo ${isSlng ? "model-card__logo--slng" : "model-card__logo--provider"}`}>
          {logoContent}
        </div>
        <div className="model-card__header-text">
          <div className="model-card__title-row">
            <span className="model-card__name">{name}</span>
            <span className={`model-card__badge ${typeBadgeClass}`}>{typeLabel}</span>
          </div>
          <div className="model-card__subtitle">{subtitle}</div>
        </div>
      </div>

      {description && <p className="model-card__description">{description}</p>}

      {bestFor && <p className="model-card__best-for"><strong>Best for:</strong> {bestFor}</p>}

      {features.length > 0 && <div className="model-card__pills">
          {features.map((f, i) => <span key={i} className="model-card__pill">{f}</span>)}
        </div>}

      <div className="model-card__footer">
        {regions.length > 0 && <div className="model-card__regions">
            <svg className="model-card__pin" xmlns="http://www.w3.org/2000/svg" width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" strokeWidth="2" strokeLinecap="round" strokeLinejoin="round">
              <path d="M20 10c0 6-8 12-8 12s-8-6-8-12a8 8 0 0 1 16 0Z" />
              <circle cx="12" cy="10" r="3" />
            </svg>
            {displayRegions}
          </div>}
        {isSlng && <span className="model-card__hosting model-card__hosting--slng">
            {hostingLabel}
          </span>}
      </div>
    </div>;
  if (href) {
    return <a href={href} className="model-card__link">{cardContent}</a>;
  }
  return cardContent;
};

## SLNG-hosted

Models hosted and optimized on SLNG infrastructure for the lowest latency.

<CardGroup cols={2}>
  <ModelCard name="Rime Arcana v3" type="tts" provider="Rime" hosting="slng" description="Rime's latest model with natural prosody across 4 languages." features={["4 languages", "Low latency"]} regions={["ap-southeast-2", "asia-south1", "eu-north-1"]} bestFor="Conversational agents, multilingual apps" href="/api-reference/tts/rime-arcana-v3/arcana-v3-english-ws" />

  <ModelCard name="Rime Arcana v1" type="tts" provider="Rime" hosting="slng" description="150+ voices across 5 languages including Arabic." features={["5 languages", "Low latency"]} regions={["ap-southeast-2"]} bestFor="High-volume production, broad language coverage" href="/api-reference/tts/rime-arcana-v2/arcana-v2-english-ws" />

  <ModelCard name="Deepgram Aura 2" type="tts" provider="Deepgram" hosting="slng" description="Deepgram Aura 2 hosted on SLNG with English and Spanish endpoints." features={["2 languages", "Low latency"]} regions={["eu-north-1", "us-central1"]} bestFor="Low-latency voice apps, multilingual IVR" href="/api-reference/tts/deepgram-aura-2/aura-2-english-ws" />

  <ModelCard name="Inworld Max 1.5" type="tts" provider="Inworld" hosting="slng" description="SLNG-hosted Inworld TTS 1.5 Max for multilingual, expressive synthesis over HTTP and bidirectional streaming WebSocket." features={["15 languages", "Expressive"]} regions={["nebius-eu-north1"]} bestFor="Expressive multilingual voice agents" href="/api-reference/tts/inworld-max/inworld-max-ws" />
</CardGroup>

## Cartesia

<CardGroup cols={2}>
  <ModelCard name="Cartesia Sonic 3" type="tts" provider="Cartesia" hosting="Cartesia" description="Sonic 3 delivers high-quality, natural-sounding speech with fine-grained generation controls including speed, volume, and emotion" features={[]} regions={["eu", "in", "us"]} bestFor="Voice agents, Voice Assistants, Real time voice calls, Fast calls" href="/api-reference/tts/cartesia-sonic-3/cartesia-sonic-3-ws" />
</CardGroup>

## Deepgram

Direct proxy to Deepgram's TTS API.

<CardGroup cols={2}>
  <ModelCard name="Aura 2" type="tts" provider="Deepgram" hosting="Deepgram" description="Deepgram's flagship TTS model via direct API proxy. 68 voices across 10 languages." features={["7 languages", "Medium latency"]} regions={["eu", "us"]} bestFor="Deepgram-native workflows" href="/api-reference/tts/deepgram-aura-2/aura-2-ws" />
</CardGroup>

## Gradium

<CardGroup cols={2}>
  <ModelCard name="Gradium TTS" type="tts" provider="Gradium" hosting="Gradium" description="Real-time multilingual text-to-speech with streaming WebSocket and one-shot HTTP synthesis." features={["5 languages", "Real-time streaming", "Low latency"]} regions={["eu"]} bestFor="Voice agents, real-time voice apps, low-latency streaming" href="/api-reference/tts/gradium-tts/gradium-tts-ws" />
</CardGroup>

## KugelAudio

Multilingual TTS with expressiveness control.

<CardGroup cols={2}>
  <ModelCard name="Kugel 1 Turbo" type="tts" provider="KugelAudio" hosting="KugelAudio" description="87 voices with expressiveness control across 26 languages." features={["26 languages", "Low latency"]} regions={["eu"]} bestFor="Multilingual voice agents, conversational AI" href="/api-reference/tts/kugel-1-turbo/kugel-1-turbo-ws" />

  <ModelCard name="Kugel 1" type="tts" provider="KugelAudio" hosting="KugelAudio" description="87 voices with expressiveness control across 26 languages. Higher quality, higher latency than Turbo." features={["26 languages", "Medium latency"]} regions={["eu"]} bestFor="Multilingual content creation, expressive speech" href="/api-reference/tts/kugel-1/kugel-1-ws" />

  <ModelCard name="Kugel 2" type="tts" provider="KugelAudio" hosting="KugelAudio" description="87 voices with expressiveness control across 26 languages. Higher quality, higher latency than Turbo." features={["26 languages", "Medium latency"]} regions={["eu"]} bestFor="Multilingual content creation, expressive speech" href="/api-reference/tts/kugel-2/kugel-2-ws" />

  <ModelCard name="Kugel 2 Turbo" type="tts" provider="KugelAudio" hosting="KugelAudio" description="87 voices with expressiveness control across 26 languages. Faster, lower-latency variant of Kugel 2." features={["26 languages", "Low latency"]} regions={["eu"]} bestFor="Multilingual voice agents, low-latency expressive speech" href="/api-reference/tts/kugel-2-turbo/kugel-2-turbo-ws" />
</CardGroup>

## Murf

<CardGroup cols={2}>
  <ModelCard name="Murf Falcon" type="tts" provider="Murf" hosting="Murf" description="Murf Falcon real-time TTS over WebSocket." features={[]} regions={["ae", "au", "eu", "gb", "in", "jp", "kr", "us"]} bestFor="Voice Agents, Low latency use cases, Cost effective, Assistants, IVR" href="/api-reference/tts/murf-falcon/murf-falcon-ws" />
</CardGroup>

## Sarvam AI

Indian-language specialist.

<CardGroup cols={2}>
  <ModelCard name="Bulbul v3" type="tts" provider="Sarvam AI" hosting="Sarvam" description="Purpose-built for Indian languages with 30+ speaker voices." features={["11 languages", "Medium latency"]} regions={["in"]} bestFor="India-focused voice apps, regional language support" href="/api-reference/tts/sarvam-ai-bulbul-v3/bulbul-v3-ws" />
</CardGroup>

## Soniox

<CardGroup cols={2}>
  <ModelCard name="Soniox TTS RT v1" type="tts" provider="Soniox" hosting="Soniox" description="Real-time text-to-speech with streaming WebSocket and one-shot HTTP synthesis." features={["Real-time streaming", "Low latency"]} regions={["eu", "jp", "us"]} bestFor="Voice agents, real-time voice apps, low-latency streaming" href="/api-reference/tts/soniox-tts-v1/soniox-tts-v1-ws" />
</CardGroup>