WOLFRAM

Enable JavaScript to interact with content and submit forms on Wolfram websites. Learn how
Wolfram Language & System Documentation Center

SpeechSynthesize [expr]

synthesizes the contents of expr as an Audio object.

SpeechSynthesize [expr,voice]

uses the specified voice to synthesize the speech signal.

Details and Options
Details and Options Details and Options
Examples  
Basic Examples  
Scope  
Options  
GeneratedAssetLocation  
Language  
Method  
OverwriteTarget  
Properties & Relations  
See Also
Related Guides
History
Cite this Page

SpeechSynthesize [expr]

synthesizes the contents of expr as an Audio object.

SpeechSynthesize [expr,voice]

uses the specified voice to synthesize the speech signal.

Details and Options

  • Speech synthesis is also known as text to speech and attempts to produce naturally worded speech rather than literal representations of expression structure.
  • SpeechSynthesize ["string"] synthesizes the text in "string".
  • SpeechSynthesize [expr] works with mathematical expressions, graphics and other constructs.
  • Use SpokenString [expr] to get the text of the speech generated by SpeechSynthesize [expr].
  • SpeechSynthesize [expr] synthesizes using a default voice. VoiceStyleData lists all available voices.
  • SpeechSynthesize accepts all SpokenString options with the following additions: [List of all options]
  • GeneratedAssetFormat Automatic the format of the result
    GeneratedAssetLocation $GeneratedAssetLocation location of the generated asset
    Language "English" the language to use for synthesis
    Method Automatic the method to use
    OverwriteTarget False whether to overwrite an existing file
  • Possible method settings include:
  • Automatic local synthesizer (default)
    TemplateBox[{"ElevenLabs", paclet:ref/service/ElevenLabs}, StringTypeLink, BaseStyle -> {2ColumnTableMod}] ElevenLabs synthesizer
    "GoogleSpeech" Google speech synthesizer
    "NeuralNetwork" synthesis using a local neural network
    "OperatingSystem" using the operating system (macOS and Windows only)
  • The "ElevenLabs" and "GoogleSpeech" methods rely on the user providing their own authentication credentials.
  • By default, the synthesized speech is placed under the "Audio" directory in $WolframDocumentsDirectory .
  • Using Language->Automatic , the language of "string" is identified using LanguageIdentify .
  • SpeechSynthesize uses machine learning. Its methods, training sets and biases included therein may change and yield varied results in different versions of the Wolfram Language.

    • List of all options

    • "ArraySizeLimit" Infinity maximal number of elements of an array or arguments of a function
      "DetailedGraphics" False controls the description of graphics primitives
      "ExpressionDepthLimit" Infinity maximal depth of an expression allowed in the output
      GeneratedAssetFormat Automatic the format of the result
      GeneratedAssetLocation $GeneratedAssetLocation location of the generated asset
      "IncludeParentheses" False controls the delimiting of subexpressions
      "IntegerLimit" Infinity maximal integer that will be given with all its digits
      Language "English" the language to use for synthesis
      Method Automatic the method to use
      OverwriteTarget False whether to overwrite an existing file
      "PostProcess" True whether to remove special formatting used by Speak from the output
      "RealPrecisionLimit" 3 how many significant digits to include in approximate numbers
      "StringLengthLimit" Infinity maximal number of characters in strings and symbol names

Examples

open all close all

Basic Examples  (2)

Speech synthesis of a piece of text:

Speech-synthesized spoken form of an expression:

Scope  (5)

Speech synthesis using a specific voice:

Use $VoiceStyles to get the list of available voices:

Synthesize using a female voice:

Synthesize using an Italian voice:

Synthesize using a voice specified by an Association :

Options  (5)

GeneratedAssetLocation  (1)

Specify the location for the generated asset:

Use GeneratedAssetLocation None to create an in-core Audio object:

Language  (2)

Synthesize an Italian sentence:

Synthesize an English sentence using an Italian accent:

Translate an English text to Italian before synthesizing:

Use Language Automatic to automatically detect the language and use an appropriate voice, when available.

Method  (1)

By default, a locally available synthesizer is used:

Use a local neural network:

Use the GoogleSpeech API:

Use ElevenLabs:

OverwriteTarget  (1)

By default, a file will not be overwritten if it already exists:

Overwrite the target file:

Properties & Relations  (1)

Use Speak to immediately start speaking an expression:

Use AudioPlay to play the synthesized audio:

Wolfram Research (2017), SpeechSynthesize, Wolfram Language function, https://reference.wolfram.com/language/ref/SpeechSynthesize.html (updated 2024).

Text

Wolfram Research (2017), SpeechSynthesize, Wolfram Language function, https://reference.wolfram.com/language/ref/SpeechSynthesize.html (updated 2024).

CMS

Wolfram Language. 2017. "SpeechSynthesize." Wolfram Language & System Documentation Center. Wolfram Research. Last Modified 2024. https://reference.wolfram.com/language/ref/SpeechSynthesize.html.

APA

Wolfram Language. (2017). SpeechSynthesize. Wolfram Language & System Documentation Center. Retrieved from https://reference.wolfram.com/language/ref/SpeechSynthesize.html

BibTeX

@misc{reference.wolfram_2025_speechsynthesize, author="Wolfram Research", title="{SpeechSynthesize}", year="2024", howpublished="\url{https://reference.wolfram.com/language/ref/SpeechSynthesize.html}", note=[Accessed: 16-November-2025]}

BibLaTeX

@online{reference.wolfram_2025_speechsynthesize, organization={Wolfram Research}, title={SpeechSynthesize}, year={2024}, url={https://reference.wolfram.com/language/ref/SpeechSynthesize.html}, note=[Accessed: 16-November-2025]}
Top [フレーム]

AltStyle によって変換されたページ (->オリジナル) /