norbertgrzenkowicz/llama_v2v

Voice to Voice Llama 3 responder

stars

commits

Python

primary language

Aug 5, 2024

updated

README

#llama_v2v

Voice to voice Llama3 responder with openai-whisper and MARS5-TTS as audio-to-text and text-to-audio models with RunPod serverless GPU computing via REST API.

In the end project looks good, but pushing to image to docker takes ages, computing on edge takes ages. It's boring. I'm abandoning the ship. If you'd like to test it, build the image and start with glue.doStuff in notebook.

Possible next chapter is transforming it from serverless REST API to wild notebook so it can run on free GPU on colab or kaggle :)))

Contributors

norbertgrzenkowicz

6 commits

Similar repos

lalanikarim/webrtc-ai-voice-chat

A WebRTC server that allows you to interact with an LLM using your speech and responds back with…

145

youjelly/lulumarjaan

b40deep/ekigambo

Adding (local language) voice prompts and audio responses to LLMs

lhl/voicechat2

Local SRT/LLM/TTS Voicechat

780

Lex-au/Orpheus-FastAPI

High-performance Text-to-Speech server with OpenAI-compatible API, 8 voices, emotion tags, and…

717

awtestergit/SpeakingAI

SpeakingAI is a demo of privately deployable 'GPT-4o like AI + RAG', a fully functional web AI…

fluxions-ai/vui

Real-time voice assistant — WebRTC streaming, faster-whisper ASR, local LLM, Vui Nano (300M) TTS.…

760

JienanZhu/speech-to-speech-minilive

speech to sppech定制的存档（数字人桥接 dh_bridge.py、豆包 TTS、DH_live 集成等）

Languages

Python

93.8%

Dockerfile

6.2%

README

#llama_v2v

Voice to voice Llama3 responder with openai-whisper and MARS5-TTS as audio-to-text and text-to-audio models with RunPod serverless GPU computing via REST API.

Possible next chapter is transforming it from serverless REST API to wild notebook so it can run on free GPU on colab or kaggle :)))