Custom Serbian Voice Elevenlabs issue

Yesterday i integrated ElevenLabs API for Serbian language, but in appears as a custom language in english language not in Serbian section. How to fix it ?

Hello,

After checking, we can confirm that the selected voice is currently identified as an English voice. That is why it is displayed under the English language category as a custom language instead of appearing in the Serbian section.

image

Here is another voice (Serbian) but still remains only in English section.

Could you please select Croatian? What do you see there?

Yeah, but for example this voice is also appear in Russian, German, Italian sections but not in Serbian section

It appears in Croatian section (same as with English,German,Italian sections etc. but then it has an accent, so i need it appears in Serbian section)

Could you please add this prompt to the core description? I’ve informed my teammates, and they will investigate the issue.

You can speak in the following languages: [‘sr-RS’]. You should only respond user in one of above languages, depending on the user’s query.

Yes i used this promt, but the accent still remains, so i need it be in that serbian section. Please let me know when this issue resolved

Hi @Mountain

We checked and confirmed, that Serbian is only available via the elevenlabs’ v3 model. ( Ref : https://help.elevenlabs.io/hc/en-us/articles/13313366263441-What-languages-do-you-support )

But we use elevenlabs’ flash 2.5 since it best fits our use case.
Due to this reason, the voice didn’t appear under Serbian.

But since its a private voice, we can enable the voice to use elevenlabs’ v3 model. Let us know if thats what you want, we can make the required updates.

Yes please make for me this version

image

Hi @Mountain

you should now have access to the voice under the Serbian language.

Hello, yes i see it on avaialable voices, but it doesn’t work at all. Just silient when i try to speak

Can you share the character-id you are trying with?

238a4646-3c96-11f1-957b-42010a7be02c

Hi Mountain

Apparently eleven_v3 model doesn’t support streaming tts and we use this for out tts system. Hence it was throwing server rejected WebSocket connection: HTTP 403" error.

Right now you wont be able to use the voice with the Serbian language.
Let me revert back the settings for the voice, so that you can use it with other available languages.

Okay, please revert it back