Open Abe-Telo opened 10 months ago
Hey! Actually I'm not sure if that XML is correct, looks like it's wss https but I believe it should just be wss
Make sure the remote host doesn't start with a protocol like the readme example
Hey! Actually I'm not sure if that XML is correct, looks like it's wss https but I believe it should just be wss
Yeah, I realized that during the test, and removed the https, But it still did not get it to work, Also as for default. I replaced that link with my link.
I will show more logs soon.
Hey! Actually I'm not sure if that XML is correct, looks like it's wss https but I believe it should just be wss
Yeah, I realized that during the test, and removed the https, But it still did not get it to work, Also as for default. I replaced that link with my link.
I will show more logs soon.
I'm also facing a similar issue like this
Have you got any fixes?
Hm interesting, at some point I was able run as is but wouldn't be crazy if something broke. Currently travelling but I'll try to debug/clean up repo in the next week.
Added some example scripts to the README that seem to work fine for me.
127.0.0.1 - - [2023-09-18 17:14:16] "POST /incoming-voice HTTP/1.1" 200 289 0.000662 INFO:pyngrok.process.ngrok:t=2023-09-18T17:14:16+0530 lvl=info msg="join connections" obj=join id=e38116dff61e l=127.0.0.1:8080 r=54.242.188.146:50054 INFO:pyngrok.process.ngrok:t=2023-09-18T17:14:18+0530 lvl=info msg="join connections" obj=join id=88bb1700f2cd l=127.0.0.1:8080 r=34.203.254.230:49818 INFO:root:Call connected, {'accountSid': '', 'streamSid': '', 'called': '', 'tracks': ['inbound'], 'mediaFormat': {'encoding': 'audio/x-mulaw', 'sampleRate': 8000, 'channels': 1}} -> Hello! Welcome to the Machine Learning hotline, how can I help? ['Hello! Welcome to the Machine Learning hotline, how can I help?'] Exception in thread Thread-8 (run_chat): Traceback (most recent call last): File "/usr/local/Cellar/python@3.11/3.11.5/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 1038, in _bootstrap_inner self.run() File "/usr/local/Cellar/python@3.11/3.11.5/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 975, in run self._target(*self._args, **self._kwargs) File "/Users/saravanarajapp/development/ai/llm_convo/twilio_ngrok_ml_rhyme_hotline.py", line 39, in run_chat run_conversation(agent_a, agent_b) File "/Users/saravanarajapp/development/ai/llm_convo/llm_convo/conversation.py", line 10, in run_conversation text_b = agent_b.get_response(transcript) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/saravanarajapp/development/ai/llm_convo/llm_convo/agents.py", line 64, in get_response self._say(transcript[-1]) File "/Users/saravanarajapp/development/ai/llm_convo/llm_convo/agents.py", line 59, in _say duration = self.speaker.get_duration(tts_fn) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/saravanarajapp/development/ai/llm_convo/llm_convo/audio_output.py", line 46, in get_duration duration = float(output.split("=")[1].split("\r")[0]) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ValueError: could not convert string to float: '4.944000\n[/FORMAT]\n' INFO:root:Call media stream ended.
Thanks for the example scripts. I've tried that keyboard chat with GPT that works fine but the other examples with twilio integration throw me the above errors when I connect the call and the API is triggered.
Hm whats your ffmpeg -version
seems like its very close to working but its unable to do some audio conversions
my ffmpeg version is 6.0
I just reinstalled ffmpeg+ffprobe (realized it's ffprobe
thats behind the issue):
ffprobe version 6.0-essentials_build-www.gyan.dev Copyright (c) 2007-2023 the FFmpeg developers
built with gcc 12.2.0 (Rev10, Built by MSYS2 project)
configuration: --enable-gpl --enable-version3 --enable-static --disable-w32threads --disable-autodetect --enable-fontconfig --enable-iconv --enable-gnutls --enable-libxml2 --enable-gmp --enable-bzlib --enable-lzma --enable-zlib --enable-libsrt --enable-libssh --enable-libzmq --enable-avisynth --enable-sdl2 --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxvid --enable-libaom --enable-libopenjpeg --enable-libvpx --enable-mediafoundation --enable-libass --enable-libfreetype --enable-libfribidi --enable-libvidstab --enable-libvmaf --enable-libzimg --enable-amf --enable-cuda-llvm --enable-cuvid --enable-ffnvcodec --enable-nvdec --enable-nvenc --enable-d3d11va --enable-dxva2 --enable-libvpl --enable-libgme --enable-libopenmpt --enable-libopencore-amrwb --enable-libmp3lame --enable-libtheora --enable-libvo-amrwbenc --enable-libgsm --enable-libopencore-amrnb --enable-libopus --enable-libspeex --enable-libvorbis --enable-librubberband
libavutil 58. 2.100 / 58. 2.100
libavcodec 60. 3.100 / 60. 3.100
libavformat 60. 3.100 / 60. 3.100
libavdevice 60. 1.100 / 60. 1.100
libavfilter 9. 3.100 / 9. 3.100
libswscale 7. 1.100 / 7. 1.100
libswresample 4. 10.100 / 4. 10.100
libpostproc 57. 1.100 / 57. 1.100
python examples\twilio_ngrok_ml_rhyme_hotline.py --preload_whisper --start_ngrok
And it seems to work 🤔
Ah think I got it, was windows vs linux issue with \r
: https://github.com/sshh12/llm_convo/commit/ba19dad208583fae181a41f9aedb76781222db78
Try updating the repo and trying it.
I encoutered the same error as above with the get_duration function, despite the last code update. It seems that ffprobe is to blame. On my AWS EC2 instance it wasn't enough to install ffmpeg and ffprobe the traditional way, but I also had to symlink ffprobe into my virtual environment. Then it worked flawlessly.
ln -sfn /usr/local/bin/ffmpeg/ffmpeg-*-amd64-static/ffprobe /path/to/your/venv/bin/ffprobe
i tried to add some debugs, but none are hitting. Is there a requirement that we are missing in the setup?
I have also added
And i see the XML is returning correctly.
What can be the issue here?