Sound update docs

This commit is contained in:
Frank 2021-06-05 23:40:50 +02:00
parent 5a00f461e9
commit cf83abfe90
2 changed files with 78 additions and 20 deletions

View File

@ -4,7 +4,7 @@
--
-- ## Features:
--
-- * Managed Radio Transmissions.
-- * Manage Radio Transmissions
--
-- ===
--
@ -15,6 +15,10 @@
--- Manages radio transmissions.
--
-- The main goal of the RADIOQUEUE class is to string together multiple sound files to play a complete sentence.
-- The underlying problem is that radio transmissions in DCS are not queued but played "on top" of each other.
-- Therefore, to achive the goal, it is vital to know the precise duration how long it takes to play the sound file.
--
-- @type RADIOQUEUE
-- @field #string ClassName Name of the class "RADIOQUEUE".
-- @field #boolean Debugmode Debug mode. More info.

View File

@ -5,7 +5,7 @@
-- ## Features:
--
-- * Create a SOUNDFILE object (mp3 or ogg) to be played via DCS or SRS transmissions
-- * Create a SOUNDTEXT object for text-to-speech output
-- * Create a SOUNDTEXT object for text-to-speech output vis SRS Simple-Text-To-Speech
--
-- ===
--
@ -13,6 +13,12 @@
--
-- ===
--
-- There are two classes, SOUNDFILE and SOUNDTEXT, defined in this section that deal with playing
-- sound files or arbitrary text (via SRS Simple-Text-To-Speech), respectively.
--
-- The SOUNDFILE and SOUNDTEXT objects can be defined and used in other MOOSE classes.
--
--
-- @module Sound.SoundOutput
-- @image Sound_SoundOutput.png
@ -63,9 +69,39 @@ do -- Sound File
--- Sound files used by other classes.
--
-- # 1. USERFLAG constructor
-- # The SOUNDFILE Concept
--
-- * @{#USERFLAG.New}(): Creates a new USERFLAG object.
-- A SOUNDFILE object hold the important properties that are necessary to play the sound file, e.g. its file name, path, duration.
--
-- It can be created with the @{#SOUNDFILE.New}(*FileName*, *Path*, *Duration*) function:
--
-- local soundfile=SOUNDFILE:New("My Soundfile.ogg", "Sound File/", 3.5)
--
-- ## SRS
--
-- If sound files are supposed to be played via SRS, you need to use the @{#SOUNDFILE.SetPlayWithSRS}() function.
--
-- # Location/Path
--
-- ## DCS
--
-- DCS can only play sound files that are located inside the mission (.miz) file. In particular, DCS cannot make use of files that are stored on
-- your hard drive.
--
-- The default location where sound files are stored in DCS is the directory "l10n/DEFAULT/". This is where sound files are placed, if they are
-- added via the mission editor (TRIGGERS-->ACTIONS-->SOUND TO ALL). Note however, that sound files which are not added with a trigger command,
-- will be deleted each time the mission is saved! Therefore, this directory is not ideal to be used especially if many sound files are to
-- be included since for each file a trigger action needs to be created. Which is cumbersome, to say the least.
--
-- The recommended way is to create a new folder inside the mission (.miz) file (a miz file is essentially zip file and can be opened, e.g., with 7-Zip)
-- and to place the sound files in there. Sound files in these folders are not wiped out by DCS on the next save.
--
-- ## SRS
--
-- SRS sound files need to be located on your local drive (not inside the miz). Therefore, you need to specify the full path.
--
--
-- ## SRS
--
-- @field #SOUNDFILE
SOUNDFILE={
@ -178,11 +214,11 @@ do -- Sound File
return name
end
--- Get the complete sound file name inlcuding its path.
--- Set whether sound files should be played via SRS.
-- @param #SOUNDFILE self
-- @param #boolean Switch If true or nil, use SRS. If false, use DCS transmission.
-- @return #SOUNDFILE self
function SOUNDFILE:UseSRS(Switch)
function SOUNDFILE:SetPlayWithSRS(Switch)
if Switch==true or Switch==nil then
self.useSRS=true
else
@ -207,12 +243,30 @@ do -- Text-To-Speech
--- Text-to-speech objects for other classes.
--
-- # Constructor
-- # The SOUNDTEXT Concept
--
-- A SOUNDTEXT object holds all necessary information to play a general text via SRS Simple-Text-To-Speech.
--
-- It can be created with the @{#SOUNDTEXT.New}(*Text*, *Duration*) function.
--
-- * @{#SOUNDTEXT.New}(*Text, Duration*): Creates a new SOUNDTEXT object.
--
-- # Options
--
-- ## Gender
--
-- You can choose a gender ("male" or "femal") with the @{#SOUNDTEXT.SetGender}(*Gender*) function.
-- Note that the gender voice needs to be installed on your windows machine for the used culture (see below).
--
-- ## Culture
--
-- You can choose a "culture" (accent) with the @{#SOUNDTEXT.SetCulture}(*Culture*) function, where the default (SRS) culture is "en-GB".
--
-- Other examples for culture are: "en-US" (US accent), "de-DE" (German), "it-IT" (Italian), "ru-RU" (Russian), "zh-CN" (Chinese).
--
-- Note that the chosen culture needs to be installed on your windows machine.
--
-- # Specific Voice
-- ## Specific Voice
--
-- You can use a specific voice for the transmission with the @{SOUNDTEXT.SetVoice}(*VoiceName*) function. Here are some examples
--
@ -287,8 +341,19 @@ do -- Text-To-Speech
return self
end
--- Set TTS culture - local for the voice.
-- @param #SOUNDTEXT self
-- @param #string Culture TTS culture. Default "en-GB".
-- @return #SOUNDTEXT self
function SOUNDTEXT:SetCulture(Culture)
self.culture=Culture or "en-GB"
return self
end
--- Set to use a specific voice name.
-- See the list from `DCS-SR-ExternalAudio.exe --help` or if using google see https://cloud.google.com/text-to-speech/docs/voices
-- See the list from `DCS-SR-ExternalAudio.exe --help` or if using google see [google voices](https://cloud.google.com/text-to-speech/docs/voices).
-- @param #SOUNDTEXT self
-- @param #string VoiceName Voice name. Note that this will overrule `Gender` and `Culture`.
-- @return #SOUNDTEXT self
@ -299,15 +364,4 @@ do -- Text-To-Speech
return self
end
--- Set TTS culture - local for the voice.
-- @param #SOUNDTEXT self
-- @param #string Culture TTS culture. Default "en-GB".
-- @return #SOUNDTEXT self
function SOUNDTEXT:SetCulture(Culture)
self.culture=Culture or "en-GB"
return self
end
end