Skip to content
Sign in
  • Browse all toolsFilter by category and pricing
  • Browse categoriesWriting, code, image, audio and more
  • Compare toolsPut up to three side by side
  • Read the blogGuides and comparisons

Unreal Speech

Text-to-speech API with audio streaming, long-form synthesis, and word-level timestamps.

Unreal Speech is a text-to-speech API that converts text into streamed audio or longer synthesized recordings, with adjustable voice settings and word-level timestamps. Unreal Speech serves developers building speech and listening applications, with access through a free API key, a one-time free allowance, and volume-based discounts.

Updated 8 Nov 20266 sections

Unreal Speech website screenshot

What is Unreal Speech?

Unreal Speech is a text-to-speech API for developers adding spoken audio to applications. Unreal Speech supports immediate audio streaming, synchronous speech generation, and asynchronous processing of longer text. The website demonstrates fiction, non-fiction, news, blog, and conversation content, and provides a live demo alongside API documentation and implementation examples.

Unreal Speech combines audio generation with timing information for applications that need text and speech to stay aligned. Supported endpoints can return word-level or sentence-level timestamps, while a dedicated WebSocket endpoint streams audio with word timing. Access begins with a free API key, and the website lists a one-time free usage allowance.

What Unreal Speech can do

Unreal Speech provides different endpoints for short audio streams, synchronous requests, and longer synthesis jobs. Request settings include voice selection, bitrate, speed, and pitch. Timestamp support varies by endpoint, allowing developers to choose between raw audio streaming, audio and timestamp URLs, or a WebSocket connection that carries audio and word timing together.

  • Unreal Speech streams raw audio through /stream.
  • Unreal Speech returns MP3 and JSON timestamp URLs through /speech.
  • Unreal Speech processes up to 500,000 characters through /synthesisTasks.
  • Unreal Speech streams audio and word-level timestamps through /streamWithTimestamps.
  • Unreal Speech supports word-level and sentence-level timestamp requests.
  • Unreal Speech accepts an optional callback URL for completed synthesis tasks.

How Unreal Speech works

Unreal Speech HTTP examples authenticate requests with a bearer API key and send text, a voice identifier, and audio settings. The /stream endpoint returns raw audio, while /speech returns audio and timestamp URLs. The /synthesisTasks endpoint returns a task identifier that developers can use in a subsequent request to check processing status.

Unreal Speech also documents a WebSocket workflow for receiving audio and timestamps during synthesis. The examples collect binary audio chunks separately from JSON timing messages, then save the resulting audio and timestamps. Python, JavaScript, and React Native examples illustrate request handling, audio playback, or output storage, depending on the example.

Who Unreal Speech is for

Unreal Speech is suited to developers building applications that turn written material into listening experiences. The demonstrated content types include fiction, non-fiction, news, blogs, and conversation. Word timing supports synchronized text highlighting, while the separate short-request and long-form endpoints accommodate applications with different text lengths and processing workflows.

  • Unreal Speech can generate spoken versions of fiction and non-fiction.
  • Unreal Speech can convert news and blog text into audio.
  • Unreal Speech can supply timestamps for word-by-word highlighting.
  • Unreal Speech can add streamed speech to React Native applications.

Unreal Speech pricing and access

Unreal Speech lists a one-time free allowance of 1M characters at $0, with an estimated audio duration of approximately 22 hours. The website offers a free API key and advertises discounts as usage increases. A high-volume inquiry option and a contact route for custom solutions are also available.

Unreal Speech strengths and limitations

Unreal Speech provides distinct workflows for raw streaming, timestamped audio, and long-form synthesis, with code examples showing how to handle each response type. Endpoint limits remain important when selecting a workflow: short streaming requests do not include timestamps, and long-form generation uses asynchronous task handling rather than an immediate completed audio response.

  • Unreal Speech limits /stream requests to 1,000 characters and returns no timestamps.
  • Unreal Speech limits /speech requests to 3,000 characters.
  • Unreal Speech returns a task identifier for asynchronous /synthesisTasks requests.

Unreal Speech FAQ

What is Unreal Speech?

Unreal Speech is a text-to-speech API that converts written text into audio. Unreal Speech supports raw audio streaming, synchronous speech generation, asynchronous long-form synthesis, and timestamp output for applications that synchronize spoken audio with text.

Is Unreal Speech free?

Unreal Speech offers a free API key and a one-time allowance of 1M characters at $0, estimated at approximately 22 hours of audio. Unreal Speech also advertises volume discounts and provides a high-volume inquiry option for custom requirements.

Does Unreal Speech support word-level timestamps?

Unreal Speech supports word-level timestamps through /speech, /synthesisTasks, and the /streamWithTimestamps WebSocket endpoint. Unreal Speech also supports sentence-level timestamps on /speech and /synthesisTasks, while the basic /stream endpoint returns raw audio without timestamps.

How much text can Unreal Speech process per request?

Unreal Speech documents limits of 1,000 characters for /stream, 3,000 characters for /speech, and 500,000 characters for /synthesisTasks. Unreal Speech processes long-form synthesis asynchronously and returns a task identifier for checking status.

Does Unreal Speech provide code examples?

Unreal Speech provides Python, JavaScript, and React Native code examples. Unreal Speech examples cover authenticated API requests, audio streaming, timestamp retrieval, asynchronous task status checks, WebSocket handling, and audio playback in a React Native application.

From the community

No reviews yet

Be the first to review this tool. What you write helps the next person choose.

More tools like Unreal Speech

VibeVoice

www.github.com

Open-source speech recognition and real-time text-to-speech models for voice AI research

Free

nobodywho

www.github.com

Open-source inference engine for local LLMs, tool calling, speech recognition and speech synthesis

Free

Beepbooply

www.beepbooply.com

AI text to speech with 900+ voices across 80+ languages and adjustable pacing, pitch, volume, and speaking styles.

Freemium

ReadSpeaker

www.readspeaker.com

AI text-to-speech for education, enterprise applications, accessible content and custom brand voices.

Paid