TTS-1-HD

Text-to-Speech • OpenAI • Proxied

OpenAI's high-definition text-to-speech model producing higher quality audio output.

Model Info
Terms and License	link ↗
More information	link ↗
Pricing	View pricing in the Cloudflare dashboard ↗

Usage

const response = await env.AI.run(
  'openai/tts-1-hd',
  {
    text: 'Hello! Welcome to Cloudflare AI Gateway. Let me show you what we can do.',
    voice: 'alloy',
    response_format: 'mp3',
    speed: 1,
  },
  {
    gateway: { id: 'default' },
  }
)
console.log(response)

Output
Raw response

{
  "state": "Completed",
  "result": {
    "audio": "https://pub-04a6d208d361438ea01b797e6973bd19.r2.dev/catalog/openai__tts-1-hd/simple-speech.mp3"
  },
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

Examples

Storytelling — HD narration with the Fable voice

const response = await env.AI.run(
  'openai/tts-1-hd',
  {
    text: 'Once upon a time, in a kingdom beyond the clouds, there lived a young inventor who dreamed of building machines that could think.',
    voice: 'fable',
    response_format: 'mp3',
    speed: 0.9,
  },
  {
    gateway: { id: 'default' },
  }
)
console.log(response)

Output
Raw response

{
  "state": "Completed",
  "result": {
    "audio": "https://pub-04a6d208d361438ea01b797e6973bd19.r2.dev/catalog/openai__tts-1-hd/storytelling.mp3"
  },
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

Podcast Style — Conversational podcast narration

const response = await env.AI.run(
  'openai/tts-1-hd',
  {
    text: "So here's the thing about large language models — they're not actually thinking. They're predicting the next token based on patterns in their training data. But the results can be surprisingly coherent.",
    voice: 'echo',
    response_format: 'mp3',
    speed: 1,
  },
  {
    gateway: { id: 'default' },
  }
)
console.log(response)

Output
Raw response

{
  "state": "Completed",
  "result": {
    "audio": "https://pub-04a6d208d361438ea01b797e6973bd19.r2.dev/catalog/openai__tts-1-hd/podcast-style.mp3"
  },
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

Shimmer Voice — Bright and expressive voice

const response = await env.AI.run(
  'openai/tts-1-hd',
  {
    text: 'Breaking news: scientists have discovered a new species of deep-sea fish that produces its own light using bioluminescence.',
    voice: 'shimmer',
    response_format: 'mp3',
    speed: 1,
  },
  {
    gateway: { id: 'default' },
  }
)
console.log(response)

Output
Raw response

{
  "state": "Completed",
  "result": {
    "audio": "https://pub-04a6d208d361438ea01b797e6973bd19.r2.dev/catalog/openai__tts-1-hd/shimmer-voice.mp3"
  },
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

text

stringrequiredmaxLength: 4096The text to generate audio for. Maximum length is 4096 characters.

voice

stringrequireddefault: alloyenum: alloy, echo, fable, onyx, nova, shimmerThe voice to use when generating the audio. Defaults to alloy.

response_format

stringrequireddefault: mp3enum: mp3, opus, wav, aac, flacThe output format for the audio. Supported formats are mp3, opus, wav, aac and flac.

speed

numberrequireddefault: 1minimum: 0.25maximum: 4The speed of the generated audio. Select a value from 0.25 to 4.0. 1.0 is the default.

audio

stringURL to the generated audio file

API Schemas (Raw)

Input

Output