SpeechGPT

SpeechGPT copied to clipboard

Timeline to release the code of SpeechGPT-Gen ?

1

KuofengGao

s2u_dir="speechgpt/utils/speech2unit/"; vocoder_dir="speechgpt/utils/vocoder/"; python3 speechgpt/src/infer/cli_infer.py --model-name-or-path "SpeechGPT-7B-cm" --lora-weights "SpeechGPT-7B-com" --s2u-dir "${s2u_dir}" --vocoder-dir "${vocoder_dir}" --output-dir "output/" ===================================BUG REPORT=================================== Welcome to bitsandbytes. For bug reports, please run python -m bitsandbytes and submit this...

rrscholarship

I am unable to run the code using conda environment. I tried running in different pytorch images I have in my server but the issue persists. Whenever I give the...

shahaamirbader

French is the language, France is the Country. It doesn't look good as the first example in the paper.

GameSetAndMatch

mandarin support？

3

Does this model support mandarin?

zdj97

See PR #22

pablorubinstein

README file has a typo in the mhubert install steps. Corrected it: `uitls` -> `utils`

pablorubinstein

Hi, This project looks really cool! Would you please consider adding a license, such as MIT or Apache 2.0? Thank you so much!

fakerybakery

Is there a chance that you may train a model with a larger context capacity like the Llama-2? Thanks!

rohan1561

``` --vocoder-dir "${vocoder_dir}" ```

showlibia