Eva

[fasca]

Bonjour,

pourriez vous ajouter plus de détails dans vos instructions s'il vous plaît ?

Instructions Expose OPENAI_API_KEY env variable in your shell Copy .env.sample to .env ##EST CE QUE RENOMMER LE FICHIER EST SUFFISANT? create bin folder in root directory and add required binaries : whisper -> main program from whisper.cpp ##SI ON INSTALLE VIA "pip install -U openai-whisper" sox ##QU'EST CE QUE C'EST? at least one ggml whisper model (I use quantized french fine-tunes from here) ##UNE FOIS LE REPO CLONé, IL FAUT QUANTIZER QUEL FICHIER AVEC QUELLE COMMANDE? Mac M1 binaries at your own risk Expect many bugs and hacks pnpm i ##QU'EST CE QUE C'EST? pnpm dev ##QU'EST CE QUE C'EST?

Merci d'avance pour votre aide. Bonne journée;

Est-ce que vous avez prévu une version conteneurisé? ça serait plus facile pour tester :)

[etienno83]

Hello @fasca, j'ai tapé les instructions du Read.me sur chatGPT, et voila le résultat : Based on the instructions provided in the "README" file of the "eva" project, it seems like the project requires you to perform the following steps to set up and run it on your local machine:

1. Expose OPENAI_API_KEY env variable: You need to set up an environment variable named OPENAI_API_KEY in your shell. This variable is likely required for API access to the OpenAI service.

2. Copy .env.sample to .env: There is a sample .env file provided in the repository. You should copy the content of this sample file into a new file named .env and adjust the settings as needed. This file might contain various environment configurations for the project.

3. Create a "bin" folder and add required binaries: You need to create a folder named "bin" in the root directory of the project. Inside this folder, you should add the following binaries:

   • whisper: This seems to be the main program, possibly compiled from the whisper.cpp source file.
   • sox: This is likely the SoX (Sound eXchange) tool, which may be used for audio processing.
   • At least one ggml whisper model: This project uses pre-trained ggml (Generic Game-Modelling Library) whisper models. You can find these models from the provided link or elsewhere and place them in the "bin" folder.

   For Mac M1 users, there might be some specific binaries or configurations required, as mentioned in the instructions.

4. Install dependencies: You need to install project dependencies using pnpm (package manager for Node.js). Run the following command in the root directory of the project:

   pnpm i

5. Run the development server: Once the dependencies are installed, you can run the development server using the following command:

   pnpm dev

Please note that the project maintainer warns about potential bugs and hacks, so you might encounter issues during the setup and running of the project.

Remember, without access to the actual code, I can't provide a complete review or resolve any specific issues. If you encounter any problems during the setup or have further questions about the codebase, feel free to ask, and I'll do my best to assist you.