TTS Models - Mint Starter Kit

Overview

sub200 provides unified access to state-of-the-art open-source TTS models through WebSocket API. Our curated selection covers edge deployment to research-grade synthesis.

Model Playground

Try models in your browser

Pricing

Simple, transparent pricing

Available Models

Enterprise Models
Edge Models

Maya Research - Maya One

State-of-the-art research model with advanced neural architecture.

Specifications

Parameters: Enterprise-grade
Latency: ~400ms
Quality: 4.9/5.0
GPU Required: Yes

Best for: Research applications, premium audiobooks, high-end content production

{
  "model": "maya-research/maya1",
  "type": "research",
  "quality": "studio"
}

Canopy Labs - Orpheus Three Billion

Professional model with three billion parameters for high-quality synthesis.

Specifications

Parameters: 3B
Latency: ~350ms
Quality: 4.8/5.0
GPU Required: Yes

Best for: Commercial production, media content, professional narration

{
  "model": "canopylabs/orpheus-3b-0.1-ft",
  "type": "professional",
  "parameters": "3B"
}

Nari Labs - Dia One Point Six Billion

Versatile conversational model with natural dialogue capabilities.

Specifications

Parameters: 1.6B
Latency: ~220ms
Quality: 4.5/5.0
GPU Required: Yes

Best for: Conversational AI, virtual assistants, interactive applications

{
  "model": "nari-labs/Dia-1.6B",
  "type": "conversational",
  "parameters": "1.6B"
}

Sesame - CSM One Billion

Balanced billion-parameter model for production use.

Specifications

Parameters: 1B
Latency: ~180ms
Quality: 4.4/5.0
GPU Required: Yes

Best for: General purpose applications, stable performance, production systems

{
  "model": "sesame/csm-1b",
  "type": "balanced",
  "parameters": "1B"
}

Hexgrad - Kokoro Eighty-Two Million

Ultra-lightweight model for edge deployment.

Specifications

Parameters: 82M
Latency: ~30ms
Memory: Less than 500MB
GPU Required: No

Best for: IoT devices, mobile applications, real-time systems

{
  "model": "hexgrad/Kokoro-82M",
  "type": "edge",
  "size": "330MB"
}

Neuphonic - NeuTTS Air

Cloud-optimized lightweight model.

Specifications

Parameters: Compact
Latency: ~75ms
Throughput: High
GPU Required: Optional

Best for: API services, cloud applications, high-volume processing

{
  "model": "neuphonic/neutts-air",
  "type": "cloud",
  "optimized": true
}

ResembleAI - Chatterbox

Dialogue-optimized conversational model.

Specifications

Parameters: Medium
Latency: ~150ms
Emotions: Six types
GPU Required: Recommended

Best for: Chatbots, NPCs, interactive dialogue systems

{
  "model": "ResembleAI/chatterbox",
  "type": "dialogue",
  "emotions": ["neutral", "happy", "sad", "angry", "fearful", "surprised"]
}

Coqui - XTTS Version Two

Cross-lingual synthesis with voice cloning.

Specifications

Languages: Twenty-five plus
Latency: ~200ms
Voice Cloning: Yes
GPU Required: Yes

Best for: Multilingual applications, dubbing, voice adaptation

{
  "model": "coqui/XTTS-v2",
  "type": "multilingual",
  "languages": 25,
  "cloning": true
}

Model Comparison

Model	Parameters	Latency	Quality	GPU	Use Case
maya-research/maya1	Enterprise	~400ms	4.9/5	Yes	Research
canopylabs/orpheus-3b-0.1-ft	3B	~350ms	4.8/5	Yes	Production
nari-labs/Dia-1.6B	1.6B	~220ms	4.5/5	Yes	Dialogue
sesame/csm-1b	1B	~180ms	4.4/5	Yes	Balanced
hexgrad/Kokoro-82M	82M	~30ms	4.0/5	No	Edge
neuphonic/neutts-air	Compact	~75ms	4.2/5	Optional	Cloud
ResembleAI/chatterbox	Medium	~150ms	4.3/5	Recommended	Interactive
coqui/XTTS-v2	Large	~200ms	4.5/5	Yes	Multilingual

WebSocket API

Connect to any model through our unified WebSocket API endpoint.

const WebSocket = require('ws');

const ws = new WebSocket('wss://api.sub200.dev/v1/tts', {
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY'
  }
});

ws.on('open', () => {
  ws.send(JSON.stringify({
    model: 'hexgrad/Kokoro-82M',
    text: 'Hello, world!',
    voice_settings: {
      stability: 0.75,
      similarity_boost: 0.75,
      style: 0.5,
      speed: 1.0
    },
    output_format: 'mp3_44100_128'
  }));
});

ws.on('message', (data) => {
  // Handle audio chunks
  console.log('Received audio chunk:', data.length, 'bytes');
});

import websocket
import json

def on_message(ws, message):
    print(f"Received audio chunk: {len(message)} bytes")

def on_open(ws):
    request = {
        "model": "hexgrad/Kokoro-82M",
        "text": "Hello, world!",
        "voice_settings": {
            "stability": 0.75,
            "similarity_boost": 0.75,
            "style": 0.5,
            "speed": 1.0
        },
        "output_format": "mp3_44100_128"
    }
    ws.send(json.dumps(request))

ws = websocket.WebSocketApp(
    "wss://api.sub200.dev/v1/tts",
    header={"Authorization": "Bearer YOUR_API_KEY"},
    on_open=on_open,
    on_message=on_message
)

ws.run_forever()

package main

import (
    "github.com/gorilla/websocket"
    "encoding/json"
)

func main() {
    headers := map[string][]string{
        "Authorization": {"Bearer YOUR_API_KEY"},
    }
    
    c, _, _ := websocket.DefaultDialer.Dial(
        "wss://api.sub200.dev/v1/tts", 
        headers
    )
    defer c.Close()
    
    request := map[string]interface{}{
        "model": "hexgrad/Kokoro-82M",
        "text": "Hello, world!",
        "voice_settings": map[string]interface{}{
            "stability": 0.75,
            "similarity_boost": 0.75,
            "style": 0.5,
            "speed": 1.0,
        },
        "output_format": "mp3_44100_128",
    }
    
    c.WriteJSON(request)
}

Request Parameters

string

required

Model identifier for synthesisAvailable models:

maya-research/maya1
hexgrad/Kokoro-82M
neuphonic/neutts-air
coqui/XTTS-v2
ResembleAI/chatterbox
sesame/csm-1b
nari-labs/Dia-1.6B
canopylabs/orpheus-3b-0.1-ft

string

required

Text to synthesize (max length varies by model and plan)

object

Voice customization parameters

number

default:"0.75"

Voice consistency (0.0 to 1.0)

number

default:"0.75"

Voice character strength (0.0 to 1.0)

number

default:"0.5"

Speaking style intensity (0.0 to 1.0)

number

default:"1.0"

Speech rate multiplier (0.5 to 2.0)

number

default:"0"

Pitch shift in semitones (-12 to 12)

string

default:"mp3_44100_128"

Audio output formatOptions:

mp3_44100_128: MP3 at 44.1kHz, 128kbps
mp3_22050_32: MP3 at 22.05kHz, 32kbps
pcm_16000: Raw PCM at 16kHz
pcm_22050: Raw PCM at 22.05kHz
pcm_44100: Raw PCM at 44.1kHz

Response Format

string

Base64-encoded audio chunk

object

Stream metadata (sent with first chunk)

number

Total audio duration in milliseconds

number

Audio sample rate

number

Number of audio channels

boolean

Indicates if streaming is complete

Model Selection Guide

By Use Case

Real-time Applications

Recommended: hexgrad/Kokoro-82M or neuphonic/neutts-air
Latency: 30-75ms
Use for: Voice assistants, live translation

Professional Content

Recommended: maya-research/maya1 or canopylabs/orpheus-3b-0.1-ft
Quality: Studio-grade
Use for: Audiobooks, podcasts, narration

Conversational AI

Recommended: ResembleAI/chatterbox or nari-labs/Dia-1.6B
Features: Emotion, context awareness
Use for: Chatbots, virtual assistants

Multilingual Projects

Recommended: coqui/XTTS-v2
Languages: Twenty-five plus
Use for: International apps, dubbing

By Performance

Lowest Latency

hexgrad/Kokoro-82M: ~30ms
neuphonic/neutts-air: ~75ms
ResembleAI/chatterbox: ~150ms

Highest Quality

maya-research/maya1: 4.9/5.0
canopylabs/orpheus-3b-0.1-ft: 4.8/5.0
coqui/XTTS-v2: 4.5/5.0

Best Efficiency

hexgrad/Kokoro-82M: 82M params
neuphonic/neutts-air: Optimized
ResembleAI/chatterbox: Balanced

Advanced Features

Voice Cloning
Emotion Control
SSML Support

Available with coqui/XTTS-v2 model.

const request = {
  model: "coqui/XTTS-v2",
  text: "Hello with cloned voice",
  voice_clone: {
    audio_url: "https://example.com/sample.wav",
    language: "en"
  }
};

Requirements:

Ten to thirty seconds of clean audio
Single speaker only
Sixteen kHz or higher sample rate

Available with conversational models.

const request = {
  model: "ResembleAI/chatterbox",
  text: "I'm excited!",
  voice_settings: {
    emotion: "excited",
    emotion_strength: 0.8
  }
};

Available emotions:

neutral
happy
sad
angry
fearful
surprised

Use SSML markup for fine control.

<speak>
  <prosody rate="slow" pitch="+2st">
    Slower and higher pitched.
  </prosody>
  <break time="500ms"/>
  <emphasis level="strong">
    Emphasized text!
  </emphasis>
</speak>

Supported tags:

break: Add pauses
emphasis: Add stress
prosody: Control rate, pitch, volume
say-as: Pronunciation hints

Error Codes

Code	Description	Resolution
INVALID_API_KEY	Invalid API key	Check credentials
RATE_LIMIT_EXCEEDED	Too many requests	Upgrade or wait
MODEL_NOT_FOUND	Invalid model ID	Check model name
TEXT_TOO_LONG	Exceeds limit	Split text
INVALID_PARAMETERS	Bad request	Check format
INSUFFICIENT_CREDITS	No credits	Add credits
SERVER_ERROR	Internal error	Retry

Best Practices

Text Preprocessing

import re

def preprocess_text(text):
    # Expand abbreviations
    replacements = {
        "Dr.": "Doctor",
        "Mr.": "Mister",
        "Mrs.": "Misses",
        "vs.": "versus"
    }
    
    for abbr, full in replacements.items():
        text = text.replace(abbr, full)
    
    # Clean special characters
    text = re.sub(r'[^\w\s.,!?;:\-\'"()]', '', text)
    
    # Ensure proper ending punctuation
    if not text.endswith(('.', '!', '?')):
        text += '.'
    
    return text

Connection Management

class ConnectionManager {
  constructor(apiKey) {
    this.apiKey = apiKey;
    this.reconnectAttempts = 0;
    this.maxReconnects = 5;
  }

  connect() {
    this.ws = new WebSocket('wss://api.sub200.dev/v1/tts', {
      headers: { 'Authorization': `Bearer ${this.apiKey}` }
    });

    this.ws.on('close', () => {
      this.handleReconnect();
    });
  }

  handleReconnect() {
    if (this.reconnectAttempts < this.maxReconnects) {
      const delay = Math.pow(2, this.reconnectAttempts) * 1000;
      setTimeout(() => {
        this.reconnectAttempts++;
        this.connect();
      }, delay);
    }
  }
}

Error Handling

class ErrorHandler:
    def handle_error(self, error):
        error_handlers = {
            'RATE_LIMIT_EXCEEDED': self.handle_rate_limit,
            'INVALID_API_KEY': self.handle_auth_error,
            'TEXT_TOO_LONG': self.handle_text_length,
            'SERVER_ERROR': self.handle_server_error
        }
        
        handler = error_handlers.get(
            error['code'], 
            self.handle_generic
        )
        return handler(error)
    
    def handle_rate_limit(self, error):
        reset_time = error['details']['reset_at']
        print(f"Rate limit hit. Resets at {reset_time}")
        # Implement backoff strategy

SDK Installation

npm install @sub200/tts

pip install sub200

go get github.com/sub200/go-sdk

cargo add sub200

Migration Guide

From ElevenLabs
From Google TTS
From Amazon Polly

Model Mapping:

eleven_multilingual_v2 → canopylabs/orpheus-3b-0.1-ft
eleven_turbo_v2 → neuphonic/neutts-air
eleven_flash_v2 → hexgrad/Kokoro-82M

Key Differences:

WebSocket vs REST API
Real-time streaming by default
Open-source models
More granular control

Support

Documentation

Comprehensive guides and tutorials

Discord

Join our community

GitHub

SDKs and examples

Status

Service monitoring

Need help? Contact our support team at sumit@sub200.dev

sub200 - Democratizing voice AI with open-source models

​Overview

Model Playground

Pricing

​Available Models

​Maya Research - Maya One

​Canopy Labs - Orpheus Three Billion

​Nari Labs - Dia One Point Six Billion

​Sesame - CSM One Billion

​Hexgrad - Kokoro Eighty-Two Million

​Neuphonic - NeuTTS Air

​ResembleAI - Chatterbox

​Coqui - XTTS Version Two

​Model Comparison

​WebSocket API

​Request Parameters

​Response Format

​Model Selection Guide

​Advanced Features

​Error Codes

​Best Practices

​SDK Installation

​Migration Guide

​Support

Documentation

Discord

GitHub

Status

Overview

Available Models

Maya Research - Maya One

Canopy Labs - Orpheus Three Billion

Nari Labs - Dia One Point Six Billion

Sesame - CSM One Billion

Hexgrad - Kokoro Eighty-Two Million

Neuphonic - NeuTTS Air

ResembleAI - Chatterbox

Coqui - XTTS Version Two

Model Comparison

WebSocket API

Request Parameters

Response Format

Model Selection Guide

Advanced Features

Error Codes

Best Practices

SDK Installation

Migration Guide

Support