Skip to content
Cloudflare Docs
DirectoryAPISDKsChangelog
Inworld logo

Inworld TTS 2

Text-to-Speech • Inworld • Proxied

Inworld's most powerful and expressive text-to-speech model. Builds on TTS 1.5 with rich expressive speech, real-time latency, natural language steering (e.g. [whisper], [say excitedly]), and stronger multilingual support across 15 production languages plus 90+ experimental languages.

Model Info
Terms and Licenselink ↗
More informationlink ↗
PricingView pricing in the Cloudflare dashboard ↗

Usage

TypeScript
const response = await env.AI.run(
  'inworld/tts-2',
  {
    output_format: 'mp3',
    temperature: 1,
    text: 'Hello! Welcome to Cloudflare AI Gateway. Let me show you what we can do.',
    timestamp_type: 'none',
    voice_id: 'Dennis',
  },
)
console.log(response)

Examples

Natural Language Steering — Direct the voice with bracketed natural-language cues for emotion, pace, and style.
TypeScript
const response = await env.AI.run(
  'inworld/tts-2',
  {
    output_format: 'mp3',
    temperature: 1,
    text: "[speak with excitement] I'm really excited about Inworld's new model. Have you tried out the steering capabilities? It's pretty cool!",
    timestamp_type: 'none',
    voice_id: 'Dennis',
  },
)
console.log(response)
Whisper — Use steering tags to whisper
TypeScript
const response = await env.AI.run(
  'inworld/tts-2',
  {
    output_format: 'mp3',
    temperature: 1,
    text: '[whisper] This is a secret just between us.',
    timestamp_type: 'none',
    voice_id: 'Dennis',
  },
)
console.log(response)
High Quality Audio — Higher sample rate for studio quality
TypeScript
const response = await env.AI.run(
  'inworld/tts-2',
  {
    output_format: 'mp3',
    sample_rate: 48000,
    temperature: 1,
    text: 'This recording is generated at studio quality for the best possible listening experience.',
    timestamp_type: 'none',
    voice_id: 'Dennis',
  },
)
console.log(response)
With Text Normalization — Expand numbers and abbreviations before synthesis
TypeScript
const response = await env.AI.run(
  'inworld/tts-2',
  {
    apply_text_normalization: true,
    output_format: 'mp3',
    temperature: 1,
    text: 'The meeting is at 3:30 PM on Jan 15th, 2026. Please confirm by calling 555-0123.',
    timestamp_type: 'none',
    voice_id: 'Dennis',
  },
)
console.log(response)

Parameters

apply_text_normalization
booleanWhen enabled, text normalization expands numbers, dates, times, and abbreviations before converting to speech. Turning this off may reduce latency.
bit_rate
integermaximum: 9007199254740991minimum: -9007199254740991Bits per second of the audio. Only for compressed audio formats (mp3, opus). The default is 128,000.
output_format
stringrequireddefault: mp3enum: mp3, opus, wav, flacThe output format for the audio. Supported formats are mp3, opus, wav, and flac. Defaults to mp3.
sample_rate
integermaximum: 9007199254740991minimum: -9007199254740991The synthesis sample rate in hertz. Accepts: 8000, 16000, 22050, 24000, 32000, 44100, 48000. The default is 48,000.
speaking_rate
numbermaximum: 1.5minimum: 0.5Speaking rate/speed, in the range [0.5, 1.5]. The default is 1.0. We recommend using values above 0.8 to ensure high quality.
temperature
numberrequireddefault: 1maximum: 2minimum: 0.01Determines the degree of randomness when sampling audio tokens. Defaults to 1.0. Accepts values between 0 (exclusive) and 2 (inclusive). Higher values = more expressive, lower values = more deterministic.
text
stringrequiredmaxLength: 2000The text to be synthesized into speech. Maximum input of 2,000 characters.
timestamp_type
stringrequireddefault: noneenum: none, word, characterControls timestamp metadata returned with the audio. "word" returns word-level timing, "character" returns character-level timing. Note: adds latency. Defaults to none.
voice_id
stringrequireddefault: Dennisenum: Loretta, Darlene, Marlene, Hank, Evelyn, Celeste, Pippa, Tessa, Liam, Callum, Hamish, Abby, Graham, Rupert, Mortimer, Snik, Anjali, Saanvi, Arjun, Claire, Oliver, Simon, Elliot, James, Serena, Gareth, Vinny, Lauren, Jessica, Ethan, Tyler, Jason, Chloe, Veronica, Victoria, Miranda, Sebastian, Victor, Malcolm, Nate, Brian, Amina, Kelsey, Derek, Evan, Kayla, Jake, Grant, Tristan, Nadia, Selene, Marcus, Riley, Damon, Cedric, Mia, Naomi, Jonah, Levi, Avery, Brandon, Conrad, Bianca, Lucian, Trevor, Alex, Ashley, Craig, Deborah, Dennis, Edward, Elizabeth, Hades, Julia, Pixie, Mark, Olivia, Priya, Ronald, Sarah, Shaun, Theodore, Timothy, Wendy, Dominus, Hana, Clive, Carter, Blake, Luna, Reed, Duncan, Felix, Eleanor, SophieThe ID of the voice to use for synthesizing speech. Defaults to Dennis.

API Schemas (Raw)

Input
Output