Add new voices

[oscie57]

I am sure there are TTS voices for other languages, so maybe try and figure out the options?

• [x] English
• [ ] Arabic
• [x] German
• [x] Spanish
• [ ] Suomi
• [x] French
• [ ] French (CA)
• [x] Indonesian
• [x] Japanese
• [x] Korean
• [ ] Malay
• [ ] Russian
• [ ] Thai
• [ ] Tieng Viet
• [ ] Chinese Simp
• [ ] Chinese Trad
• [ ] Hebrew
• [ ] Cebuano
• [ ] Cestina
• [ ] Italian
• [ ] Magyar
• [ ] Netherlandish
• [ ] Polish
• [ ] Romanian
• [ ] Svenska
• [ ] Filipino
• [ ] Greek
• [ ] Ukrainain
• [ ] Hindi

These are all the languages supported in the app.

[gitlimes]

I don't think TTS is available for all of the languages, though. Like, I tried everything from it_000 to it_030 (and also it_it_000 to it_it_030) and all gave me the same error (The currently used timbre has been offline). Might write a script or something to test it

[gitlimes]

Alright, here's the voices i found with the script (tested 000-046 at time of writing):

br_001     Brazilian F1
br_003     Brazilian F2
br_004     Brazilian F3
br_005     Brazilian F4
de_001     German F1
de_002     German F2
en_au_001  Australian English F
en_au_002  Australian English M
en_uk_001  British English M1
en_uk_003  British English M2
en_us_001  American English F1
en_us_002  American English F1 (duplicate)
en_us_006  American English M1
en_us_007  American English M2
en_us_009  American English M3
en_us_010  American English M4
es_002     Spanish M
es_mx_002  Mexican Spanish M
fr_001     French M1
fr_002     French M2
id_001     Indonesian F
jp_001     Japanese F1
jp_003     Japanese F2
jp_005     Japanese F3
jp_006     Japanese M
kr_002     Korean M1
kr_003     Korean F
kr_004     Korean M2

[oscie57]

I don't think TTS is available for all of the languages, though. Like, I tried everything from it_000 to it_030 (and also it_it_000 to it_it_030) and all gave me the same error (The currently used timbre has been offline). Might write a script or something to test it

me and a friend found out that italy doesnt have access to the tts so it makes sense

[gitlimes]

I don't think TTS is available for all of the languages, though. Like, I tried everything from it_000 to it_030 (and also it_it_000 to it_it_030) and all gave me the same error (The currently used timbre has been offline). Might write a script or something to test it

me and a friend found out that italy doesnt have access to the tts so it makes sense

Ah, I see. Oof for SpaghettiLand I guess

[oscie57]

what voice is https://vm.tiktok.com/ZMLG634wt/

[oscie57]

https://vm.tiktok.com/ZMLtudr89/ also

[bulieme]

Alto: https://www.tiktok.com/@kkbutterfly27xxxx/video/7100492686162922757

I cant remember who. https://www.tiktok.com/@scotttenormansgf/video/7100227167811915054?is_copy_url=1&is_from_webapp=v1

There's a new text to speech that can sing, Its 2 voices, Alto, and others... i cant remember who is those

EDIT: Alto, Tenor sings.

[DetachHead]

i found the narrator and singing voices #11

[oscie57]

@ashmonty wondering if you can run that script again? possibly could pick up some new voices that ive seen around

[gitlimes]

@ashmonty wondering if you can run that script again? possibly could pick up some new voices that ive seen around

Alrighty! Do keep in mind that the script only checks for languages, not for special voices

[gitlimes]

Update: found the same voices in this list

[oscie57]

ahh shit

[oscie57]

"en_us_ghostface": "Scream",
"en_male_grinch": "Trickster",
"en_male_wizard": "Magician",
"en_female_ht_f08_halloween": "Opera",
"en_female_madam_leota": "Madame Leota",
"en_male_ghosthost": "Ghost Host",
"en_male_pirate": "Pirate",
"en_female_ht_f08_glorious": "Euphoric",
"en_male_sing_funny_it_goes_up": "Hypetrain",
"en_female_samc": "Empathetic",
"en_male_cody": "Serious",
"en_female_ht_f08_wonderful_world": "Melodrama",
"en_male_m2_xhxs_m03_silly": "Quirky Time",
"en_male_funny": "Wacky",
"en_female_emotional": "Peaceful",
"en_male_m03_sunshine_soon": "Toon Beat",
"en_female_f08_warmy_breeze": "Open Mic",
"en_male_m03_lobby": "Jingle",
"en_female_f08_salut_damour": "Cottagecore",
"es_mx_002": "Warm",
"en_male_narration": "Story Teller",
"en_us_007": "Professor",
"en_us_009": "Scientist",
"en_us_010": "Confidence",
"en_us_006": "Joey",
"en_au_001": "Metro",
"en_uk_001": "Narrator",
"en_us_002": "Jessie",
"en_au_002": "Smooth"

gotten by https://github.com/Wolfhound905

[alexdo83]

@oscie57 Voice Codes vietnamese?

[Iucasmaia]

How can i get voice codes for my language?

[Iucasmaia]

"en_female_grandma": "Grandma", "en_male_joker": "Joker", "en_male_goblin": "Goblin", "en_male_grinch": "Grinch",

[MrPromotor]

does anyone know the code for the new TiktokVoice from optimus prime?

[ClaudioBo]

does anyone know the code for the new TiktokVoice from optimus prime?

I also need it too, but the Spanish (Latin American) Optimus Prime

[ClaudioBo]

Recently for Spanish tiktok I got these voices, including Optimus Prime (not english):

• es_male_m3 - Julio
• es_mx_male_transformer - Optimus Prime (Heavily bass boosted)
• es_mx_female_supermom - Super Mom
• es_female_f6 - Alejandra
• es_female_fp1 - Mariana

[ExDrill]

@ClaudioBo How are you getting these voice codes?

[DetachHead]

i'm not sure how others have been doing it but i was using https://github.com/NSExceptional/FLEXing on a jailbroken iOS device.

i can't remember the exact steps i took because i can't get it working on my current device but i just looked through all the menus and searched for known voice names as i used them in the app. that helped me find where to look for unknown voice names and from there i was able to find new ones

[chrisjp]

I didn't see anyone mention it here, you appear to be missing en_male_santa and en_male_cupid

[ClaudioBo]

@ExDrill This is how I got those Voice IDs:

Requirements

• A Android device (Rooting is NOT required)
• A throw away Tiktok account (Optionally, you may get banned if you do care, i didn't got banned yet tho')
• HTTP Toolkit both on your phone and computer

Steps

1. Install a modified Tiktok .apk from the Releases of this repository: https://github.com/Eltion/Tiktok-SSL-Pinning-Bypass (Tiktok is protected from any MITM attacks if your connection doesn't use their internal certificate, so this modified apk removes it... and the method we're about to use is based on a MITM proxy)
2. Open the installed Tiktok and login via phone number/mail (don't use any SSO i.e. Google because it will not work)
3. Install HTTP Toolkit to intercept and analyze Tiktok traffic from the Play Store and install the included certificate from the app (the app will guide you)
4. On your computer on HTTP Toolkit select "Android device via QR code" and follow the steps shown on screen
5. When your phone is connected to your computer via HTTP Toolkit, on your phone select that you only want to intercept traffic from Tiktok ('ALL APPS' button > Three dots > Disable all apps > Search and select Tiktok)
6. Go back to Tiktok, record a quick video, write any text and try to use the TTS so Tiktok will request the voice endpoint and HTTP Toolkit would grab that information and show it on your computer
7. Search for /media/api request and see the URL parameters when you find one
8. Voila!

---

Having said the above, I want to ask the following providing the next information:

Voice list endpoint

So my idea is getting all the Voices IDs of a language by only giving the Country Code to get the available voices, instead of depending on random people to replicate the steps mentioned above to get their available voices of their country.

Obviously Tiktok queries the voice list from somewhere, so I did saw all the requests intercepted on HTTP Toolkit using my steps above and found the endpoint (/effect/api/v3/effects), query parameters ('panel=speaking-voice' is to get only the voices) and headers that Tiktok sends to that endpoint, but I can't replicate it on a cURL command because I get a empty JSON response ({}) and a 403 HTTP status code

If anybody is interested and curious enough, I hope this little information that I've found helps you (and helps everyone else who uses this whole TTS service thing)

(Changing Tiktok language doesn't work, I dont know if it works by changing the phone's system language, or using a proxy... I don't know if using a proxy is possible because HTTP Toolkit is a local proxy)

(Explaination on Step 1 may be inaccurate, sorry if im stupid lol)

[chrisjp]

found the endpoint (/effect/api/v3/effects), query parameters ('panel=speaking-voice' is to get only the voices) and headers that Tiktok sends to that endpoint, but I can't replicate it on a cURL command because I get a empty JSON response ({}) and a 403 HTTP status code

You were probably doing a POST request when it only accepts GET; my cURL requests attempting to replicate it gave specific error messages regarding what data was missing. So after much trial and error I now know which params are required. However, I need to do further testing with language and region parameters (there are a total of 6 region parameters and 3 language parameters being sent with the request) to determine what each one is actually for and if it affects the JSON returned. I'm sure there must be a way to get other language's voices from a properly crafted request.

Anyway, it requires a LOT more parameters than just panel (see screenshot below). The important one being a valid access_key - which is the current problem I'm having in generating - I don't know how/where this is created. It is NOT the same as the session ID we get from login cookies. If I can figure out how to reliably craft a working cURL request I'll let you guys know, but for now it seems HTTP Toolkit will be a requirement to obtaining a valid key.

In the meantime, I have managed to obtain all the missing English voice IDs which I'll post here later - I need to do some more testing first.

Thanks again for posting those steps @ClaudioBo, it was extremely helpful.

[chrisjp]

I figured out the request for fetching voices list JSON for each language!

Bad news it requires an access_key which as far as I can tell is generated inside the app somewhere, so use of HTTP Toolkit and the modified TikTok .apk linked in @ClaudioBo's above comment is absolutely required to obtain this. Unless of course someone is willing to publicly share theirs - I've no idea what the ramifications of doing so might be but I wouldn't have thought it's a good idea unless it's from a throwaway account.

Here's some quick and dirty documentation for it based on my testing...

API endpoint

GET https://api31-normal-useast2a.tiktokv.com/effect/api/v3/effects

Required parameters

• access_key - 32 char alphanumeric string, looks like it might be an md5 hash. Seems to be generated in the app. No cookies generated by the website contain this value. You'll need HTTP Toolkit to find this.
• app_version - self-explanatory. 30.1.2 is the value I'm using, matches the modified .apk
• sdk_version - unsure which SDK this refers to, possibly something internal. 14.3.0 is a working value.
• device_id - a 19-digit numeric [0-9] ID. You can fake this. 1234567890123456789 will work.
• device_platform - Must be android. I didn't find any other working values.
• device_type - model ID of your device. My emulator uses SM-G988N. You can probably put anything you want here.
• channel - Presumably the store the app was acquired from. googleplay works here.
• panel - Panel is what TikTok calls each of its effect categories or something. You need to put speaking-voice for this.

Optional parameters

region - 2-letter uppercase country code. Defaults to US if not set. carrier_region - 2-letter uppercase country code. Defaults to US if not set. On your device this will be the country it's physically located in based on SIM card network you're connected to.

You'll need to include both of the above parameters if you want voices specific to a region. Note that although we have voice IDs for France, Germany, Spain, and South Korea; using any of their country codes will only return the default English voice options. I don't know what's happened to those original ones but they do still work if used. Also, I suspect China uses a separate API as setting CN here also returns only the default English voices.

Full URL example

https://api31-normal-useast2a.tiktokv.com/effect/api/v3/effects?access_key=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX&app_version=30.1.2&sdk_version=14.3.0&device_id=1234567890123456789&device_platform=android&device_type=SM-G988N&channel=googleplay&region=US&carrier_region=US&panel=speaking-voice you'd need a valid access_key in that for this to work, but no special headers need to be set. You could open it in your browser if you wanted to.

I hope this proves helpful to others.

[chrisjp]

Full list of voices

Names are as shown in the app for that language, additional names in parentheses are either what other languages call the voice or is suggested by the id for it.

I haven't tried every region code in the world but I did try every one listed in the original post, and combined with old voice IDs we already knew of this is what I believe is now a fully complete list.

English

Voice ID	Name
en_uk_001	Narrator (Chris)
en_uk_003	UK Male 2
en_female_emotional	Peaceful
en_au_001	Metro (Eddie)
en_au_002	Smooth (Alex)
en_us_002	Jessie
en_us_006	Joey
en_us_007	Professor
en_us_009	Scientist
en_us_010	Confidence
en_female_samc	Empathetic
en_male_cody	Serious
en_male_narration	Story Teller
en_male_funny	Wacky
en_male_jarvis	Alfred
en_male_santa_narration	Author
en_female_betty	Bae
en_female_makeup	Beauty Guru
en_female_richgirl	Bestie
en_male_cupid	Cupid
en_female_shenna	Debutante
en_male_ghosthost	Ghost Host
en_female_grandma	Grandma
en_male_ukneighbor	Lord Cringe
en_male_wizard	Magician
en_male_trevor	Marty
en_male_deadpool	Mr. GoodGuy (Deadpool)
en_male_ukbutler	Mr. Meticulous
en_male_petercullen	Optimus Prime
en_male_pirate	Pirate
en_male_santa	Santa
en_male_santa_effect	Santa (w/ effect)
en_female_pansino	Varsity
en_male_grinch	Trickster (Grinch)
en_us_ghostface	Ghostface (Scream)
en_us_chewbacca	Chewbacca (Star Wars)
en_us_c3po	C-3PO (Star Wars)
en_us_stormtrooper	Stormtrooper (Star Wars)
en_us_stitch	Stitch (Lilo & Stitch)
en_us_rocket	Rocket (Guardians of the Galaxy)
en_female_madam_leota	Madame Leota (Haunted Mansion)
en_male_sing_deep_jingle	Song: Caroler
en_male_m03_classical	Song: Classic Electric
en_female_f08_salut_damour	Song: Cottagecore (Salut d'Amour)
en_male_m2_xhxs_m03_christmas	Song: Cozy
en_female_f08_warmy_breeze	Song: Open Mic (Warmy Breeze)
en_female_ht_f08_halloween	Song: Opera (Halloween)
en_female_ht_f08_glorious	Song: Euphoric (Glorious)
en_male_sing_funny_it_goes_up	Song: Hypetrain (It Goes Up)
en_male_m03_lobby	Song: Jingle (Lobby)
en_female_ht_f08_wonderful_world	Song: Melodrama (Wonderful World)
en_female_ht_f08_newyear	Song: NYE 2023
en_male_sing_funny_thanksgiving	Song: Thanksgiving
en_male_m03_sunshine_soon	Song: Toon Beat (Sunshine Soon)
en_female_f08_twinkle	Song: Pop Lullaby
en_male_m2_xhxs_m03_silly	Song: Quirky Time

French

Voice ID	Name
fr_001	French Male 1
fr_002	French Male 2

German

Voice ID	Name
de_001	German Female
de_002	German Male

Indonesian

Voice ID	Name
id_male_darma	Darma
id_female_icha	Icha
id_female_noor	Noor
id_male_putra	Putra

Italian

Voice ID	Name
it_male_m18	Italian Male

Japanese

Voice ID	Name
jp_001	Miho (美穂)
jp_003	Keiko (恵子)
jp_005	Sakura (さくら)
jp_006	Naoki (直樹)
jp_male_osada	モリスケ (Morisuke)
jp_male_matsuo	モジャオ (Matsuo)
jp_female_machikoriiita	まちこりーた (Machikoriiita)
jp_male_matsudake	マツダ家の日常 (Matsudake)
jp_male_shuichiro	修一朗 (Shuichiro)
jp_female_rei	丸山礼 (Maruyama Rei)
jp_male_hikakin	ヒカキン (Hikakin)
jp_female_yagishaki	八木沙季 (Yagi Saki)

Korean

Voice ID	Name
kr_002	Korean Male 1
kr_004	Korean Male 2
kr_003	Korean Female

Portuguese

Voice ID	Name
br_003	Júlia
br_004	Ana
br_005	Lucas
pt_female_lhays	Lhays Macedo
pt_female_laizza	Laizza
pt_male_transformer	Optimus Prime (Portuguese)

Spanish

Voice ID	Name
es_002	Spanish Male
es_male_m3	Julio
es_female_f6	Alejandra
es_female_fp1	Mariana
es_mx_002	Álex (Warm)
es_mx_male_transformer	Optimus Prime (Mexican)
es_mx_female_supermom	Super Mamá

Duplicated voices removed from the above table

• en_us_001 = en_us_002
• br_001 = br_004
• id_001 = id_female_noor

[thetoonsofjosh]

This is a pretty awesome list but I don't see "Italian Male" (it_male_m18 is the voice code) in the list.

[chrisjp]

Thanks, I'll edit my comment to include that.

It was stated earlier in the thread that Italy didn't have access to TTS, and when I set IT as the region which fetching voice codes only the default English voices were returned. So it's strange this one exists. Do you know if it's still usable in the app by Italian users? Because if it is there must be a different API call returning additional voice codes we might not know about yet.

[ClaudioBo]

es_mx_male_transformer is now unavailable now :(

[ExDrill]

Any updates regarding voices being added?