Documentation to integrate AI voices to LG applications. The REST API is split into three endpoints: speech-to-text, text-to-text via Groq, and text-to-speech. The documentation is split into three separate pages, one for all the model descriptions with required and optional parameters, one for how to setup and use the API, and a pricing documentation.
The table below lists all the models used in the project:
| Speech-to-Text | Text-to-Speech | LLM (via Groq) |
|---|---|---|
| Google Cloud AI with Gemini | Google Cloud AI with WaveNet | Gemma2-9b-it |
| Deepgram | Deepgram | Gemma-7b-It |
| AssemblyAI | llElevenLabs | Mixtral-8x7b-32768 |
| Llama3-8b-8192 | ||
| Llama3-70b-8192 |
To get started, navigate to the API documentation below to start setting up the environment and usage of the voice-to-voice API.
Link to the API usage for voice-to-voice integrations for LG apps are here.
Link to the description of the models and costs are here.
Link to the list of all the model pricing are listed here.