search

huggingface / speech-to-speech

Speech To Speech: an effort for an open-sourced and modular GPT4-o
Apache License 2.0
2.99k stars 314 forks source link

Please help me #25

Open luojwen120 opened 3 weeks ago

luojwen120 commented 3 weeks ago

How to start the server(ubuntu 22.04):

python s2s_pipeline.py \
    --recv_host 0.0.0.0 \
    --send_host 0.0.0.0 \
    --lm_model_name microsoft/Phi-3-mini-4k-instruct \
    --init_chat_role system \
    --stt_compile_mode reduce-overhead \
    --tts_compile_mode default

How to start the client(windows10):

python listen_and_play.py --host 10.0.24.227

When I execute the client command, the server log looks normal. image

However, when I followed the instructions on the client and pressed Enter, the client and server did not give me any feedback. Why is this? image

andimarafioti commented 3 weeks ago

Hi! Sadly I don't have a windows machine to test this. Stackoverflow says to disable nnpack with this line: torch.backends.nnpack.enabled = False