Hindi Version Or Anyone Want to work on this ?

[AMEERAZAM08]

[Zengyi-Qin]

We already have a Hindi demo in demo_part2.ipynb

[AMEERAZAM08]

its not working i have already check

On Tue, Jan 2, 2024 at 11:35 PM Zengyi Qin @.***> wrote:

We already have a Hindi demo in demo_part2.ipynb image.png (view on web) https://github.com/myshell-ai/OpenVoice/assets/40556743/81ec13b1-9f5b-4d0f-87bb-a617e5432283

— Reply to this email directly, view it on GitHub https://github.com/myshell-ai/OpenVoice/issues/24#issuecomment-1874366441, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHFL55IK4JQX5D7CEEB3LFTYMRD6NAVCNFSM6AAAAABBJ6WTUWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNZUGM3DMNBUGE . You are receiving this because you authored the thread.Message ID: @.***>

[Zengyi-Qin]

Could you describe the error?

[shaimarus]

Can anyone provide a tutorial on how to do this for other languages?

Maybe someone needs a tutorial for other languages, I run it in thru docker:

docker run --gpus all -it -v /:/home some_image_with_gpu_like_nvidia_cuda_12.1.1-devel-ubuntu20.04 bash

2 git clone https://github.com/myshell-ai/OpenVoice.git 3 ls 4 cd OpenVoice/ 5 pip3.10 install -r requirements.txt 6 history

We need download checkoints in the folder (for me it '/home/shaimarus/checkpoints')

then run in python:

import os import torch import se_extractor from api import BaseSpeakerTTS, ToneColorConverter

ckpt_converter='/home/home/shaimarus/checkpoints/converter' device = 'cuda:0' output_dir='/home/home/shaimarus/checkpoints'

tone_color_converter = ToneColorConverter(f'{ckpt_converter}/config.json', device=device) tone_color_converter.load_ckpt(f'{ckpt_converter}/checkpoint.pth')

os.makedirs(output_dir, exist_ok=True)

source=f'{output_dir}/wav1.wav' #this text in audio will read target=f'{output_dir}/wav2.wav' #this voice will reproduce text in wav1.wav save_path = f'{output_dir}/output_crosslingual.wav' # output file encode_message = "@MyShell"

source_se, audio_name = se_extractor.get_se(source, tone_color_converter, vad=True)

target_se, audio_name = se_extractor.get_se(target, tone_color_converter, vad=True)

tone_color_converter.convert( audio_src_path=source, src_se=source_se, tgt_se=target_se, output_path=save_path, message=encode_message)

To be honest, I wasn’t very impressed with the result for another lang, or I am doing something wrong

[netrunner-exe]

@Zengyi-Qin, hi! I was not able to test demo_part2.ipynb, because at the stage of loading OPENAI_API_KEY I get an error - RateLimitError: Error code: 429 - {'error': {'message': 'You exceeded your current quota, please check your plan and billing details. For more information on this error, read the docs: https://platform.openai.com/docs/guides/error-codes/api-errors.', 'type': 'insufficient_quota', 'param': None, 'code': 'insufficient_quota'}}. In OpenVoice folder I created an .env file and in it I added generated fresh OpenAI API key in the format OPENAI_API_KEY='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' (this is a random example of the key, it is not valid). I also tried writing it to a file without '', but the error is still the same. Is it possible that I'm doing something wrong or do I need a subscription to work with OpenAI TTS?