From 7f6f7c636e88665ad672ff1b17b1b70a5cffec11 Mon Sep 17 00:00:00 2001 From: Fakhir Ali Date: Sun, 16 Jun 2024 17:08:03 +0500 Subject: [PATCH] Update notes --- INSTALL.md | 4 ++++ README.md | 55 +++++++++++++++++++++++++------------------------ notes/Models.md | 3 ++- notes/TODO.md | 17 --------------- 4 files changed, 34 insertions(+), 45 deletions(-) delete mode 100644 notes/TODO.md diff --git a/INSTALL.md b/INSTALL.md index 55f7644..7eceb16 100644 --- a/INSTALL.md +++ b/INSTALL.md @@ -7,3 +7,7 @@ Make sure to install it using the correct CMAKE flag(s). - [torchaudio](https://pytorch.org/get-started/locally/) - portaudio by running `sudo apt-get install portaudio19-dev ` - [onnxruntime-gpu](https://onnxruntime.ai/docs/install/) + +```bash +pip install -r requirements.txt +``` \ No newline at end of file diff --git a/README.md b/README.md index 3a273db..e6e7e90 100644 --- a/README.md +++ b/README.md @@ -15,32 +15,7 @@ Have a natural voice conversation with an LLM https://github.com/fakhirali/OpenVoiceChat/assets/32309516/88b7973d-a362-46f3-ab18-232bb59a188e - -Supports all kinds of stt, tts and llm [models](notes/Models.md). - -Supports interruptions. - -Well [abstracted](/openvoicechat/tts) apis, easy to use and [extend](notes/Adding_models.md). - -The goal is to be the open source alternative to [closed commercial implementations](notes/Competition.md) - -Some ideas are [here](notes/Ideas.md). - -[TODO](notes/TODO.md). - -Start with the [bounties](https://docs.google.com/spreadsheets/d/1d2MZTa9FKM4IHLrBs_nMuA2yuLaSY4USzdGH6vRdPbU/edit?usp=sharing) -if you want to contribute. - -[Installation](INSTALL.md). - -```shell -python main.py -``` - -[Discord](https://discord.gg/M5S2JksapH) - -## Installing Required Packages -### To install only the base packages +### pip installation ```shell pip install openvoicechat ``` @@ -56,4 +31,30 @@ similarly "piper" and "openai" can be replaced by any of the following - openai - tortoise - xtts -- transformers \ No newline at end of file +- transformers + +```shell +python main.py +``` + +[local Installation](INSTALL.md). + +### Features + +Supports practically any stt, tts and llm [model](notes/Models.md). + +Supports interruptions. + +Well [abstracted](/openvoicechat/tts) apis, easy to use and [extend](notes/Adding_models.md). + +The goal is to be the open source alternative to [closed commercial implementations](notes/Competition.md) + +Some ideas are [here](notes/Ideas.md). + +### Contributing +Start with the [bounties](https://docs.google.com/spreadsheets/d/1d2MZTa9FKM4IHLrBs_nMuA2yuLaSY4USzdGH6vRdPbU/edit?usp=sharing) +if you want to contribute. + +Roadmap [here](notes/Roadmap.md) + +[Discord](https://discord.gg/M5S2JksapH) diff --git a/notes/Models.md b/notes/Models.md index f507ac8..4a228e1 100644 --- a/notes/Models.md +++ b/notes/Models.md @@ -11,9 +11,10 @@ huggingface. [Elevenlabs](https://elevenlabs.io/) [Any model](https://huggingface.co/models?pipeline_tag=automatic-speech-recognition&sort=trending) from huggingface. [Vosk models](https://alphacephei.com/vosk/models). +[Deepgram](https://deepgram.com/product/speech-to-text) ## LLM All models compatible with llama-cpp-python e.g [llama](https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGUF) and [mistral](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF). -OpenAI's [GPTs](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) \ No newline at end of file +OpenAI's [GPTs](https://platform.openai.com/docs/guides/text-generation/chat-completions-api). \ No newline at end of file diff --git a/notes/TODO.md b/notes/TODO.md deleted file mode 100644 index be6b229..0000000 --- a/notes/TODO.md +++ /dev/null @@ -1,17 +0,0 @@ -# TODO - -| Easy to use/extend | Supports alot of models | Competitive Product | Showcase | -|--------------------|-------------------------|------------------------------|-------------| -| Documentation | Ollama | Timing | Online Demo | -| ~~Pip package~~ | Tortoise | Websockets | visualizer | -| | ~~Deepgram~~ | Turn Taking | ~~Website~~ | -| | | ~~Interruption info to LLM~~ | | -| | | Sounddevice underruns | | -| | | ~~Streaming stt~~ | | -| | | Filler words etc | | -| | | RAG | | -| | | | | -| | | | | -| | | | | -| | | | | -| | | | |