speak

Version: org.nasdanika.texttospeech 1.0.0.202006150225
origin/master 9fe4163
Usage: nsd speak [-hjsSV] [--[no-]cache] [--cache-dir=<cacheDir>]
                 [-e=<charset>] [-o=<output>] [-p=<progressOutput>]
                 [-P=<parallelism>] [-t=<timeout>] [-v=<voice>]
                 [-c=<String=String>]... [-C=URL]... [-M=<String=String>]...
                 LANGUAGE [FILE...]
Generates voice from text using Google Text-To-Speech and local cache. Text is
interpolated with context entries.
      LANGUAGE              Language and locale code, e.g. en-US. See https:
                              //cloud.google.com/text-to-speech/docs/voices for
                              a list of supported languages.
      [FILE...]             A list of files to generate speech from. File type
                              is determined by its extension - .txt for plain
                              text and .ssml for SSML (https://cloud.google.
                              com/text-to-speech/docs/ssml) unless explicitly
                              specified by the SSML option. If FILE argument is
                              not provided then text is taken from the standard
                              input and is treated as text or SSML based on the
                              SSML option value. If a file is a directory then
                              it is processed recursively generating speach
                              from .txt and .ssml files. In this case if SSML
                              option is set to true only ssml files are
                              processed, if it is set to false then only text
                              files are processed.
  -c, --context-entry=<String=String>
                            Context entries.
                            Shadow entries in contexts and mounts.
  -C, --context=URL         Context resource URL relative to the current
                              directory. YAML, JSON, or properties. In
                              properties dots are treated as key path
                              separators. Type is inferred from the content
                              type header, if it is present, or extension.
                              Contexts are composed in the order of definition,
                              later context entries shadowing the former
      --[no-]cache          Indicates whether to use cache, default is true.
      --cache-dir=<cacheDir>
                            Cache directory, default value is C:
                              \Windows\system32\config\systemprofile\.
                              nasdanika\speech-synthesizer\cache
  -e, --charset=<charset>   Input charset, default value: UTF-8
  -h, --help                Show this help message and exit.
  -j, --json                Output progress in JSON
  -M, --context-mount=<String=String>
                            MappingContext resource URL relative to the current
                              directory. YAML, JSON, or properties. In
                              properties dots are treated as key path
                              separators. Type is inferred from the content
                              type header, if it is present, or extension.
                              Mounts shadow context entries.
  -o, --output=<output>     Output file or directory for a single input (file
                              or standard input) or an output directory for
                              multiple inputs. If not provided, generated sound
                              is output to the standard output for single input
                              and to the location of inputs for multiple
                              inputs. In the latter case or if the output is a
                              directory, audio file name is formed by adding .
                              mp3 extension to the source file name. E.g. my.
                              txt -> my.txt.mp3
  -p, --progress=<progressOutput>
                            Output file for progress monitor
  -P, --parallelism=<parallelism>
                            If the value greater than one then an executor
                              service is created and injected into the context
                              to allow concurrent execution.
  -s, --[no-]ssml           Input is SSML.
  -S, --skip-unmodified     Set this option to skip generation for files which
                              were not modified since the previous generation,
                              i.e. an output audio file exists and has a
                              timestamp greater than the timestamp of the
                              source text of SSML file.
  -t, --timeout=<timeout>   If parallelism is greater than one this option
                              specifies timout in seconds awaiting completion
                              of execution. Default value is 60.
  -v, --voice=<voice>       Voice name. See https://cloud.google.
                              com/text-to-speech/docs/voices for a list of
                              supported voices.
  -V, --version             Print version information and exit.
Exit codes:
  0   Success
  1   Unhandled exception during execution
  2   Invalid input
  3   Diagnostic failed
  4   Execution failed or was cancelled, successful rollback
  5   Execution failed or was cancelled, rollback was not successful
  6   Executor service termination timed out