Speech Markdown transpiler for Elixir.
This library converts text in the Speech Markdown format to SSML for processing by Text-To-Speech APIs, etc.
This element adds an audio filter to a section or modifier. The filter syntax allows specifying a sequence of audio filter definitions separated by semicolons, each with its own comma-separated list of name=value parameters. Whitespace is ignored inside the filter definition.
(Hello)[filter:"reverse_reverb()"]
#[
filter:"
dalek(frequency=30, distortion=4);
xsynth_fixed(frequency=110);
reverb;
"
]
You would make a good Dalek.
A filter applied in a modifier replaces any filters in surrounding modifiers
or sections. In the following example, the only filter applied to the text
will be reverb
.
#[filter:"xsynth_random"]
(Poetry is an echo, asking a shadow to dance.)[filter:"reverb"]
This element adds a specific style of diction that can be simulated by text manipulation. Examples of diction styles are a "pig Latin" effect and a sibilant lisp. The diction syntax allows specifying a semicolon-separated sequence of styles. Whitespace is ignored inside the definition.
(Hello)[diction:"pig_latin"]
#[diction:"pig_latin; lisp"]
She sells seashells down by the seashore.
A diction style applied in a modifier replaces any dictions in surrounding modifiers
or sections. In the following example, the only diction applied to the text
will be pig_latin
.
#[diction:"lisp"]
(Poetry is an echo, asking a shadow to dance.)[diction:"pig_latin"]
This element specifies an emotion in which the associated text should be read. Supported emotions will vary by model, but only one emotion can be listed at a time. An optional numeric "intensity" modifier is also allowed to vary the expression of the requested emotion.
(Hello)[emotion:"angry";intensity:"2"]
#[diction:"excited"]
She sells seashells down by the seashore.
The emotion
tag follows the same override rules as the other extensions.