Higher Female Protagonist Voice

Note: Automatron patch is now available.  Please read the special installation note below.
Note: Far Harbor patch is now available.  Please read the special installation note below.

This mod is an attempt to address the fact that the voice of the actress chosen for the protagonist in the English version of Fallout 4 is on the deep end (~160 Hz) of the range for female voices worldwide (160-280 Hz).  Aside from the simple fact of it being a low voice - much lower, for example, than the female voice used in Fallout 3 - this significantly hinders one's ability to roleplay as somebody who isn't middle-aged.

Note: I should point out that this mod is tagged as "English", and this means that it is intended for the English version of Fallout 4 and, if used, will result in English dialogue.  For this mod to work in other languages, new versions must be created for those languages.

Other languages: Japanese, French, German, Spanish.
Deeper Female Protagonist Voice is here.
Male protagonist voice mods: Deeper, higher.
Companion voice mods: Curie, Piper, Cait (higher), Cait (deeper), Strong, Preston, X6-88.
Other voice mods: Female children, Super Mutants.
Other audio mods: IMPACTFUL NUKES.

The tweaks with effectively zero distortion are those highlighted in green.





Voice Types:

IN SUMMARY: The voice types below come in basically two types - Those in which the pitch / formants adjustments are identical, and those in which they are not.  When the adjustments are equal, the voice can be considered to be effectively distortion-free.  Otherwise there will be some audible distortion in varying (and specified) amounts.

Voice Type A1:

• Pitch raised 19%, formants raised 6%.  This used to be the one I used (now I use Type D1 or Type B).  The voice sounds somewhat higher and somewhat younger.
• Pro: Relatively little audible distortion.
• Con: Ultimately not a huge change in pitch.
  

Voice Type A2:

• Pitch raised 6%, formants raised 6%.  A very slightly higher voice that sounds somewhat younger.  Included mainly so that there is an option that has as little distortion as possible while also providing at least some improvement, however minimal.
• Pro: Essentially zero audible distortion.
• Con: Barely any change in pitch.
  

Voice Type A3:

• Pitch raised 3%, formants raised 3%.  A very, very slightly higher voice that sounds just a bit younger.  Basically the midpoint between Type A2 and the default voice.  The actual modification is probably too subtle for a single piece of dialogue to reveal; one would simply need to try the voice out and see if it suits their needs.
• Pro: Essentially zero audible distortion.
• Con: Barely any change in pitch.
  

Voice Type B:

• Pitch raised 9%, formants raised 9%.  A formant raise of this magnitude means the voice sounds markedly younger than the original, though the rise in pitch is minimal.
• Pro: Essentially zero audible distortion.
• Con: Very little change in pitch.  As a straight pitch-shift, this is as much as I felt could be gotten away with without making the protagonist sound too "helium"-artificial.
  

Voice Type C:

• Pitch raised 19%, formants unaltered.  The voice sounds somewhat higher but essentially the same age.
• Pro: Audible distortion isn't excessive.
• Con: Still not a lot of change in pitch.  This is about as far as I could push the process before engendering excessive distortion in a batch convert, but there definitely is more distortion here.
  

Voice Type U1:

• Pitch raised 50%, formants unaltered.  The voice is considerably higher, bringing it in line with world-average vocal frequencies, but theoretically sounding the same age.
• Pro: Considerable change in pitch.
• Con: Considerable audible distortion, probably to the point of being unusable, hence the "U".  Use at your own discretion!
  

Voice Type U2:

• Pitch raised 50%, formants raised 9%.  The voice is considerably higher, bringing it in line with world-average vocal frequencies, and also sounds much younger.
• Pro: Considerable change in pitch.
• Con: The greatest distortion among the available choices, probably to the point of being unusable, hence the "U".  Use at your own discretion!
  

Please read this disclaimer concerning the rest of the voice types:

The entire point of going beyond 9% formants is to push for a younger voice.  These voice modifications may be too far for most people, but whenever the "shojo" type player mods begin to arrive, something like this will be needed.  However, there are some important limitations which need to be kept in mind:

• In order to prevent or at least limit distortion, there will be no significant pitch changes such as 50%+.
• The voice actress is in the habit of letting her voice hover at its lowest comfortable register.  This means that the spoken dialogue quite regularly has a hoarse-sounding quality to it, and sometimes even "vocal fry".  These qualities, while not problematic if you're supposed to be middle-aged, are incompatible with the perception of a young-sounding voice, and it is effectively impossible to simply change them or edit them out.
• Since there is considerable variability in the tonal quality of the voiced dialogue across 13,000 individual lines, some lines sound more "natural" after modification than others, and the more one pushes the boundary, the more this balance tips in favor of "unnatural".  This point needs to be kept particularly in mind.
  

Voice Type D1:

• Pitch raised 12.2%, formants raised 12.2%.  For my own purposes, I have more or less settled upon this voice.  Type B is somewhat less young-sounding.
• Pro: Essentially zero audible distortion.
• Con: Younger without being particularly higher.  Also the fact that there are still hoarse / "vocal fry" moments in the dialogue tends to undermine the youthfulness.
  

Voice Type D2:

• Pitch raised 26%, formants raised 12.2%.  Same adjustment as Type D1, but with some extra pitch adjustment.
• Pro: Low-ish audible distortion.
• Con: As with all voice types that have formants not equaling pitch, there is at least some audible distortion.  Also the fact that there are still hoarse / "vocal fry" moments in the dialogue tends to undermine the youthfulness.
  

Voice Type E1:

• Pitch raised 15.6%, formants raised 15.6%.  Pushing the voice even further than D1.  Really anything more than this and the voice isn't sounding younger so much as smaller, as in little people.
• Pro: Essentially zero audible distortion.
• Con: Younger without being particularly higher, especially for the apparent youthfulness of the voice.  Also the fact that there are still hoarse / "vocal fry" moments in the dialogue tends to undermine the youthfulness.
  

Voice Type E2:

• Pitch raised 30%, formants raised 15.6%.  Same adjustment as Type E1, but with some extra pitch adjustment.
• Pro: About as young-sounding as it can get and still be somewhat usable, though the dialogue variability does mean there are many helium moments.
• Con: This modification ended up exhibiting more distortion than I was expecting.  Also the fact that there are still hoarse / "vocal fry" moments in the dialogue tends to undermine the youthfulness.
  

Audio processes designed to change pitch are inherently imprecise, especially when used in a batch capacity.  When the pitch and formants differn weirdly artificial-sounding artifacts are all but guaranteed to be a problem.  The greater the modification, the worse the artifacts.  The results I provide may be the best which can be done with hobbyist-caliber software, without manually tweaking every single specimen of nearly 13,000 files.  Professional-grade software/hardware could possibly do a better job, and may be able to render a usable result on a 50% pitch shift.

Further caveats: The voice actress had to read thousands of lines across multiple recording sessions.  Because of this, there is considerable inconsistency in her tone of voice, including what kind of age she was shooting for.  This means that tweaks to these lines of dialogue will be similarly inconsistent, occasionally resulting in, for example, more "helium" phenomena than expected.  Also, as mentioned earlier, she is in the habit of maintaining her voice at its lowest comfortable register, with the consequence that the inflections in many lines of dialogue push this boundary.  This unfortunately results in many instances of hoarse-throat and "vocal fry" dialogue.  (Google it.)  These instances are faithfully tweaked, of course, but sound far less natural on a higher-pitched and/or younger voice than they do on a middle-aged voice.

These voice replacements will make much better sense once there exist mods which enable the player to produce a female visage that looks younger than someone in their 30s.


Installation (as of 2016-3-21):

1. Follow the Fallout 4 Mod Installation guide to learn the latest method for setting up custom mods.

2. Patch 1.3 eliminated transparent support for .bsa archives, and there still is no way of creating .ba2 archives, so here are the options for installing as I understand them (copied from the mod thread).  I have not found the time to test them.

(The following examples assume you want to install "Higher Female Protagonist Voice - Type D1.bsa".  Replace this text with the .bsa filename of your intended mod.)

Option 1:

* Go to your Fallout4.ini under My Documents/My Games/Fallout4
* Under Archive section, find 'sResourceArchiveList=' without the quotes.
* Copy and paste this line and replace the existing line after 'sResourceArchiveList=':

sResourceArchiveList=Fallout4 - Voices.ba2, Fallout4 - Meshes.ba2, Fallout4 - MeshesExtra.ba2, Fallout4 - Misc.ba2, Fallout4 - Sounds.ba2, Fallout4 - Materials.ba2, Higher Female Protagonist Voice - Type D1.bsa

* Start your game and the voice should have change to your desired voice type.

Option 2: 

* If you're afraid of messing up the Fallout4.ini file, you can just place the commands below into Fallout4Custom.ini.  It will leave the original one intact and use the one in Fallout4Custom.ini while avoiding faulty ini file.

[Archive]
sResourceArchiveList=Fallout4 - Voices.ba2, Fallout4 - Meshes.ba2, Fallout4 - MeshesExtra.ba2, Fallout4 - Misc.ba2, Fallout4 - Sounds.ba2, Fallout4 - Materials.ba2, Higher Female Protagonist Voice - Type D1.bsa

* Add the name of the esp file of your chosen voice type (for example: "Higher Female Protagonist Voice - Type D1.esp") to the plugins.txt file, which is typically located in C:\Users\(your username)\AppData\Local\Fallout4.

* Extract the contents of the archive to your Fallout4\Data\ directory.

Option 3:

* Use the Bethesda Archive Extractor utility to directly extract the contents of this mod's .bsa archive to your Data directory.


Special installation note about Automatron:

As far as I can tell (and I could be mistaken) the .ba2 archive content packaged with this DLC has been given total priority, which means none of the above options will work until one has forcibly discarded what they are trying to replace.  In other words, you will probably need to do the following before the above options will work:

1: Using Bethesda Archive Extractor, Extract the contents of your DLCRobot - Voices_en.ba2 archive to your data dir.
2: Delete the PlayerVoiceFemale01 dir, located in Sound/Voice/DLCRobot.esm/
3: Hide your DLCRobot - Voices_en.ba2 file, either by moving it to a temporary location or by temporarily renaming it.
4: Now try one of the previous three options.

Revert these changes once this latest issue has been sorted out by the modding community.  Until then, be aware that if Steam re-downloads the DLCRobot - Voices_en.ba2 file, the voice may revert to vanilla.  My apologies in advance if this information is no good / outdated; too busy with hospital visits and such lately.


Special installation note about Far Harbor:

For the time being, this patch is being made available as loose files.  Mod Organizer 2 is recommended.  After installing, the same issue that exists with the Automatron DLC also evidently exists for Far Harbor, meaning similar steps may be necessary to force the mod to take priority:

1: Using Bethesda Archive Extractor, Extract the contents of your DLCCoast - Voices_en.ba2 archive to your data dir.
2: Delete the PlayerVoiceFemale01 dir, located in Sound/Voice/DLCCoast.esm/
3: Hide your DLCCoast - Voices_en.ba2 file, either by moving it to a temporary location or by temporarily renaming it.
4: Now try one of the previous three options.


Other:

I hope this mod proves of use to others.

2015-11-13: Added a simple video showcasing the subtle differences between the various voices.
2015-11-18: Added the video to the description text since the "videos" tab isn't showing for some people.
2015-11-19: Added Type A3, a requested 3% version.
2015-11-23: Added Type D1, Type D2, Type E1, Type E2 - all aimed specifically at providing a younger voice.
2015-11-30: Began uploading archived versions of the various files, in order to circumvent a rare bug in the game.