Unreal Speech Python SDK allows you to easily integrate the Unreal Speech API into your Python applications for text-to-speech (TTS) synthesis. This SDK provides convenient methods for working with the Unreal Speech API, including generating speech, managing synthesis tasks, and streaming audio.

Getting Started


You can install the Unreal Speech API Python package using pip:

pip install unrealspeech

To ensure the package functions correctly, it is necessary to install the following dependencies as well.

pip install playsound pydub simpleaudio
/streamStream audio for short, time-sensitive cases
/speechGenerate speech with options (MP3 format)
/synthesisTasksManage synthesis tasks for longer text
/synthesisTasks/TaskIdCheck the status of a synthesis task

Common Request Body Schema

PropertyTypeRequired?Default ValueAllowed Values
VoiceIdstringRequiredN/AScarlett, Liv, Dan, Will, Amy
BitratestringOptional192k16k, 32k, 48k, 64k, 128k, 192k, 256k, 320k
SpeedfloatOptional0-1.0 to 1.0
PitchfloatOptional1.00.5 to 1.5

Parameter Details

Voice ID

  • Dan: Young Male
  • Will: Mature Male
  • Scarlett: Young Female
  • Liv: Young Female
  • Amy: Mature Female


Defaults to 192k. Use lower values for low bandwidth or to reduce the transferred file size. Use higher values for higher fidelity.


Defaults to 0.


  • 0.5: makes the audio 50% faster. (i.e., 60-second audio becomes 42 seconds)
  • -0.5: makes the audio 50% slower. (i.e., 60-second audio becomes 90 seconds.)


Defaults to 1. However, on the landing page, we default male voices to 0.92 as people tend to prefer lower/deeper male voices.

Rate Limit

PlanRequests per second


Initializing the UnrealSpeechAPI
First, import the UnrealSpeechAPI class

from unrealspeech import UnrealSpeechAPI, play, save

Then, initialize the API with your API key

api_key = 'YOUR_API_KEY'
speech_api = UnrealSpeechAPI(api_key)

Generating Speech

You can generate speech by providing a text string and optional parameters

text_to_speech = "This is a sample text."
timestamp_type = "sentence"  # Choose from 'sentence' or 'word'
voice_id = "Scarlett"  # Choose the desired voice
bitrate = "192k"
speed = 0 
pitch = 1.0
audio_data = speech_api.speech(text=text_to_speech,voice_id=voice_id, bitrate=bitrate, timestamp_type=timestamp_type, speed=speed, pitch=pitch)

# Play audio

Streaming Audio

For short and time-sensitive cases, you can use the /stream endpoint to stream audio

# Stream audio
text_to_stream = "This is a short text to be synthesized."
voice_id = "Will"
timestamp_type = "sentence"  # Choose from 'sentence' or 'word'
bitrate = "192k"
speed = 0
pitch = 1.0

# Generate audio from text
audio_data = speech_api.stream(
   text=text_to_stream, voice_id=voice_id, bitrate=bitrate, timestamp_type=timestamp_type, speed, pitch)

 # Play audio

Managing Synthesis Tasks

You can manage synthesis tasks for longer text using the synthesisTasks endpoint

# Create a synthesis task
task_id = speech_api.create_synthesis_task(text="Long content", voice_id="Will", bitrate="320k", timestamp_type="word", speed=0, pitch=1.0)

# Check the task status
audio_data = speech_api.get_synthesis_task_status(task_id)

# Play audio

Downloading Audio

You can download audio by simply calling the save function

  from unrealspeech import save

  audio_data = sppech_api.speech('How to download your audio easily')

  # you can use the save function to save the audio
  save(audio_data, "output.mp3")


Contributions are welcome! If you find any issues or have suggestions for improvements, please open an issue or submit a pull request.