Voices from Narakeet.
# support
a
I need to use a spanish voice from Narakeet (https://www.narakeet.com). It seems like: 1. Narakeet is not mentioned under the voice providers in VAPI's voice library. 2. It is not possible to connect to a custom voice provider in the same way as it is possible to connect to a custom LLM, where a custom LLM is selected and a url is provided. It seems like some of the voice providers is an option for using cloned voices. I believe that this is for making a voice sound like an particular human, which is something different from the concern to get the pronouncement of the Spanish word correct. Please advice on how solve this.
v
You can continue this conversation by mentioning me in the message.
@abj I understand you're looking to use a Spanish voice from Narakeet with VAPI. While Narakeet is not directly integrated as a voice provider in VAPI's voice library, you can still use your cloned voice from 11labs, which you've synced with VAPI, to ensure correct pronunciation in Spanish. Here's what you can do: 1. Make sure your cloned voice has been trained with the correct Spanish pronunciation. This might involve providing samples in Spanish to 11labs when creating your cloned voice. 2. Once your cloned voice is trained and synced with VAPI, you can use it to pronounce Spanish words correctly. If you need a specific Spanish voice from Narakeet and it's not available in VAPI, you might have to use Narakeet's services separately and then integrate the audio files into your application. Unfortunately, VAPI does not currently support connecting to custom voice providers in the same way it does with custom LLMs. If you require further assistance or have any other questions, please let me know!
s
@abj first you can use voice provider from the available options, second You can now integrate your own Text-to-Speech (TTS) provider. For detailed API implementation, visit https://api.vapi.ai/api-json
lastly cloned voice is to upload custom voices to give a unique voice to your bots.
a
@Shubham Bajaj 'You can now integrate your own Text-to-Speech (TTS) provider' sounds like the option I am looking for. However, I could not find this mentioned in the general api documentation. Is it possible to provide a direct link to that documentation?
s
try
assistant.voice.provider='custom-voice'
. it's API only and you can check the docs here: https://api.vapi.ai/api