siinlab/darija-chatbot

TTS data and model for Moroccan Darija # Moroccan Darija Chatbot > Before using this repository, please read the License and Terms of Use sections below. ## Project Overview This project includes several components to facilitate the development and usage of a Moroccan Darija Chatbot system: ### Datasets The project provides curated datasets necessary for training the TTS models. These datasets include audio recordings and their corresponding transcriptions in Moroccan Darija. ### Trained Models Pre-trained models are available for immediate use. These models have been trained on the provided datasets and can be used to generate speech from text in Moroccan Darija. ### Finetuning Scripts For users who wish to improve or adapt the TTS models, finetuning scripts are included. These scripts allow for further training of the models on additional data or specific use cases. ### API An API is provided to enable easy integration of the TTS system into other applications. This API allows users to input text and receive the generated speech as output. ### UI A user-friendly interface is available to quickly test the TTS model. This UI allows users to input text and listen to the generated speech, making it easy to evaluate the performance of the TTS system. ## Environment This repository has a pre-built Docker image that contains all the necessary dependencies to run the TTS system. To pull the Docker image, use the following command: ```bash docker pull ghcr.io ``` The content of the repository is mounted in the `/app` directory of the Docker container. To run the Docker container, use the following command: ```bash docker run -it ghcr.io ``` From within the Docker container, run these commands to set up the TTS system: ```bash bash models/scripts/setup-tts.sh bash models/scripts/setup-embedding.sh bash models/scripts/setup-whisper-asr.sh ``` Once you have set up the dependencies, you can use the trained models through the API and/or UI. Moreover, …