use it on colab
Can you guide me on how to use it on google colab. I ran the "pip install git+https://github.com/huggingface/parler-tts.git" but it doesn't provide me with the input "prompts" like text and description of voice.
should I run the coded in entirety?
I a not a hardcore programmer and would appreciate if you can help as I am converting Sanskrit script to YouTube videos
Hey
@rameshjey
, You can run the inference code provided on the model card page directly on collab if you have installed the repository using "pip install git+https://github.com/huggingface/parler-tts.git
"
I have been testing the model using Google Colab for a month and haven't faced any issues yet.
You can switch between languages by changing the prompt (it detects the language of your text prompt and generates the audio accordingly).
Maybe you can elaborate on your problem so I can help you with the specifics.
thanks for yr response. I ran it on colab, but i am unable to identify the output mp3 file and where is is stored. attaching screen showing successful running of code segment 2 with my input in Tamil and prompt on speaker
Firstly, I believe you are using the wrong model, which, in the case of Indic languages such as Tamil, should be "ai4bharat/indic-parler-tts." Also, I hope you are aware where the downloaded files are saved in Collab (they're in the /content directory, which you can access by clicking the folder icon on the left).
Additionally, ensure you load the model only once before running the generations. You can modify this later and rerun the cell if you want.
If you are using Collab for the first time, "Change the runtime" to an available GPU(Unsubscribed users get 4 hours of free GPU a day).
Let me know if this solves your problem.