This version improves the optional reporting on long tracks, such as the Nordschleife.
It also adds some coordination with the AutoCam to decrease camera switching in some situations while AnnouncerBot is talking.
There is now a driverNames.txt that you can use to help the TTS pronounce names correctly, as well as use nicknames for drivers in that file.
Contents of driverNames.txt:
;HOWTO: Each Row should Start with the drivers online name followed by a | (pipe) ;After the pipe there should be a list of names separated by ^s (hats) ;The drivers full name should always be listed first, with any nicknames listed afterwards, in no particular order ;ASRL David Zahn|David Zahn^Zahn Kevin Lechmanski|Kevin LechManski^LechManski ;VAC Jonah Hax|Jonah Hax^Jonah^Hax neal pardoe|neal pardoe^neal
New relevant entries in the INI
reportables = NOTHING|TYRES|LAP_NUMBER|NATIONALITY|FASTEST_LAP|BALLAST|MAKE|START_POSITION
When the camera switches to a new driver AnnouncerBot has extra reportable data beyond just the driver name and their current position. The reportables list contains all the various data points that may randomly be reported. You can remove options from this list if you would like them skipped.
ballastname = _keilos
The default ballastname used to be "kg", which would often be read by the TTS at "kilograms". You can now change this value. I have included an underscore before "keilos", as it sounds better when spoken by the voice.exe.
There is new logic that attempts to limit the amount that AnnouncerBot will speak when the camera does lots of switching between drivers. Switching drivers should always result in the drivers name and position being reported, but other data should be skipped.
The V2.0 of Announcer Bot no longer uses .WAV files of any sort, so the download is much smaller than before. If you are currently using .WAV files it should technically still be compatible, but there are many things the V2.0 ABot will not say.
V2.0 should now announce the gap between drivers as a time estimate, and will also announce some basic information about the track, the layout, and weather. ABot should now understand more (but not necessarily all) tire options, and announce a drivers nationality once per session (assuming the set their nation flag).
You can now toggle/enable/disable ABot using Ctrl+F2.
This update should resolve the issue of Announcer Bot mixing male and female voices. The new INI should contains the following items:
soundsfolder = voice
voiceCmd = apps\\python\\AnnouncerBot\\voice.exe -n "Microsoft David Desktop" --mono --khz 22 -o OUTPUTFILE TEXTTOSPEECH
isVoiceTTS = 1
Those entries are all defaults. There is an "apps\python\AnnouncerBot\sounds\voice" directory that contains all the prerendered sounds, and new .WAV files (like driver names and lap times) should be created as needed by calling the voiceCmd when .WAV files do not already exist.
The new INI entries will also allow you to try new voices. I successfully tested with the "Microsoft Elsa Desktop" voice (which I had to install) by editing the voiceCmd and setting soundsfolder = elsa. If you try a new voice it will take quite a while to load the first time as all the missing .WAV files are created.
You may want to go to the "apps\python\AnnouncerBot\sounds\voice" folder and delete any .WAV files that were created using a female voice. It would probably be easiest to just remove anything newer than whatever date you first downloaded ABot.
The default "voice" was repeating information about drivers having 0 kg ballast. That has been fixed.
The V1.1 has also been updated to announce lap times during qually sessions with using the "voice" sounds folder.
If you are getting errors using the "voice" please test the voice.exe in the assettocorsa\apps\python\AnnouncerBot folder. I've had no issues using it so far on two of my Windows 10 machines, but it may not work on older Windows operating systems.