Stability-AI/stable-audio-tools issues and pull requests

#144 - loss nan in VAE training

Issue - State: open - Opened by shwj114514 2 days ago - 1 comment

#143 - FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated on 0.0.16

Issue - State: open - Opened by carlthome 3 days ago

#142 - Training Validation - When to stop?

Issue - State: open - Opened by cvillela 10 days ago

#141 - About attention mask

Issue - State: open - Opened by SuperiorDtj 12 days ago - 15 comments

#140 - Requirements Versions - Pip takes forever to resolve

Issue - State: open - Opened by zeta274 14 days ago - 1 comment

#139 - Split requirements

Pull Request - State: open - Opened by akx 30 days ago

#138 - Finetuning stable-audio-open 1.0 - Model Weights + Configs

Issue - State: closed - Opened by cvillela about 1 month ago - 2 comments

#137 - Training from Scratch Resulting in Noise

Issue - State: open - Opened by kenkalang about 1 month ago - 2 comments

#136 - About Dataset seconds_total

Issue - State: open - Opened by UESTC-Med424-JYX about 2 months ago

#135 - Batch Size?

Issue - State: open - Opened by Trimad about 2 months ago

#134 - Mask_args for Outpaintinig

Issue - State: open - Opened by JLenzy about 2 months ago - 2 comments

#133 - does it support standard v-diffusion?

Issue - State: open - Opened by GenjiB about 2 months ago

#132 - Encodec config error and Encodec training on 16kHz data

Issue - State: open - Opened by zhangchi2004 about 2 months ago

#131 - Does the model support bf16 training？

Issue - State: open - Opened by atomrun39 2 months ago - 2 comments

#130 - How to use audio inpainting?

Issue - State: open - Opened by csu-roycheng 2 months ago

#129 - Please help with stable-audio-open model

Issue - State: open - Opened by asigalov61 2 months ago

#128 - OSError: sndfile library not found

Issue - State: open - Opened by TacoByte 2 months ago - 2 comments

#127 - ModuleNotFoundError: No module named 'sklearn.preprocessing._csr_polynomial_expansion'

Issue - State: open - Opened by Yinghoo 2 months ago

#126 - Provide helpers to compile for AWS Inferentia

Issue - State: open - Opened by vdesaintpern 2 months ago

#125 - Add GitHub Actions workflow for testing dependency installation

Pull Request - State: open - Opened by carlthome 2 months ago

#124 - Add pre-commit config for black and isort

Pull Request - State: open - Opened by carlthome 2 months ago - 2 comments

#123 - Add QK LayerNorm to transformer

Pull Request - State: closed - Opened by zqevans 2 months ago

#122 - About traning speed: DiT is slower than Unet with same batch size

Issue - State: open - Opened by SuperiorDtj 2 months ago

#121 - Set pretransform to eval during diffusion training

Pull Request - State: closed - Opened by zqevans 2 months ago

#120 - Bottleneck updates

Pull Request - State: closed - Opened by zqevans 2 months ago

#119 - When resuming an unwrapped checkpoint, LR is not updated

Issue - State: open - Opened by Taikakim 3 months ago

#118 - Tensor cores on an A100

Issue - State: open - Opened by Taikakim 3 months ago

#117 - Segmentation Faults on Dataloaders

Issue - State: closed - Opened by JLenzy 3 months ago - 1 comment

#116 - Remove unused s3fs dependency

Pull Request - State: open - Opened by ggilder 3 months ago - 3 comments

#114 - Continuing on an A100 does not work in Colab

Issue - State: open - Opened by Taikakim 3 months ago - 3 comments

#113 - Audio to audio task - Dataset configuration

Issue - State: open - Opened by Jourdelune 3 months ago - 5 comments

#112 - Separated checkpoints for VAE and DiT model for stabilityai/stable-audio-open-1.0 so we can easily perform finetunning?

Issue - State: open - Opened by BingliangLi 3 months ago - 2 comments

#111 - Arthur training

Pull Request - State: open - Opened by arthurcornelio88 3 months ago

#110 - Add gradient clipping for autoencoder

Issue - State: open - Opened by nateraw 3 months ago

#109 - Figure out and fix marination and inpainting.

Pull Request - State: open - Opened by piwell 3 months ago

#108 - Is Deep Speed faster than DDP in a cluster of 4*8 A100?

Issue - State: open - Opened by BingliangLi 3 months ago

#107 - Fix Pickling Windows

Pull Request - State: open - Opened by Katehuuh 3 months ago - 1 comment

#106 - Expose Gradio `--share` and default to `False` if not set

Pull Request - State: closed - Opened by carlthome 3 months ago - 1 comment

#105 - Please explain --pretransform-ckpt-path

Issue - State: open - Opened by neops1 3 months ago - 1 comment

#104 - [update] Auto trim the audio file

Pull Request - State: open - Opened by cyberofficial 3 months ago

#101 - gradio: shouldn't we cut audiofile to seconds_total?

Issue - State: open - Opened by elgiano 3 months ago

#100 - GPU generation problem

Issue - State: open - Opened by Crunchyguy1337 3 months ago

#99 - Training - Missing key(s) in state_dict:

Issue - State: closed - Opened by Katehuuh 3 months ago - 3 comments

#98 - Add an option to do the final step on a CPU

Pull Request - State: open - Opened by vbuterin 3 months ago - 4 comments

#97 - Cannot install. Idk about "wheel" or what subprocess could be the problem

Issue - State: closed - Opened by frankleitor 3 months ago - 1 comment

#96 - Improved usability for txt2audio ui

Pull Request - State: open - Opened by LastDemon99 3 months ago

#95 - Change GPU for inference with CMD-Flag?

Issue - State: open - Opened by oliverban 3 months ago

#94 - Multi-channel data support

Issue - State: open - Opened by fmiotello 3 months ago - 1 comment

#93 - It appears that the version 0.7.4 of pedalboard is no longer available on PyPI

Issue - State: open - Opened by DJZeroAction 3 months ago - 4 comments

#92 - README: Added link to RunPod template

Pull Request - State: open - Opened by ashleykleynhans 3 months ago

#91 - How do I run stable audio tools from just command prompt?

Issue - State: open - Opened by scaruslooner 3 months ago - 2 comments

#90 - ModuleNotFoundError: No module named 'dac.nn'

Issue - State: open - Opened by Abhijitjimmi 3 months ago - 5 comments

#89 - ModuleNotFoundError: No module named 'pathtools.patterns'

Issue - State: closed - Opened by Chilluminati91 3 months ago - 1 comment

#88 - error code on trianing run PicklingError: Can't pickle <function get_custom_metadata at 0x000001DDF6501EE0>: import of module 'metadata_module' failed

Issue - State: open - Opened by JayDeeRC 3 months ago - 2 comments

#87 - Training issue - ckpt and state_dict dont seem to match

Issue - State: closed - Opened by JayDeeRC 3 months ago

#86 - Installation not completing - botocore

Issue - State: closed - Opened by gardberg 4 months ago - 1 comment

#85 - Gradio issue - Seconds total is ignored

Issue - State: open - Opened by SoftologyPro 4 months ago - 4 comments

#84 - Share issue with gradio UI

Issue - State: closed - Opened by SoftologyPro 4 months ago - 1 comment

#83 - Enhance setup.py with best practices

Pull Request - State: open - Opened by tejhande 4 months ago

#82 - Use mps device on macOS

Pull Request - State: closed - Opened by akx 4 months ago - 9 comments

#81 - Convert ckpt to pt file for inference

Issue - State: open - Opened by Tensorpunk 4 months ago - 1 comment

#80 - Init audio conditioning doesn't work with --model-half

Issue - State: open - Opened by IIPedro 4 months ago - 1 comment

#79 - macOS M1 (Apple Silicon) Support?

Issue - State: closed - Opened by yukiarimo 4 months ago

#78 - Output directory and filename for the gradio UI

Issue - State: open - Opened by SoftologyPro 4 months ago

#77 - docs: update README.md

Pull Request - State: closed - Opened by eltociear 4 months ago

#76 - torch.cuda.OutOfMemoryError: Allocation on device

Issue - State: closed - Opened by smthemex 4 months ago

#75 - A bug in RotaryEmbedding

Issue - State: closed - Opened by yukara-ikemiya 4 months ago - 2 comments

#74 - ModuleNotFoundError: No module named 'packaging' from pip install stable-audio-tools

Issue - State: open - Opened by w84death 4 months ago - 9 comments

#73 - [Bug] Segmentation fault trying to generate with `stable-audio-open-1.0`

Issue - State: closed - Opened by NanoCode012 4 months ago - 4 comments

#72 - Windows gradio UI issue - ImportError: attempted relative import with no known parent package

Issue - State: closed - Opened by SoftologyPro 4 months ago - 8 comments

#71 - How to generate audio variations and style transfer of audio samples?

Issue - State: open - Opened by Gitterman69 4 months ago - 2 comments

#70 - Update to 0.0.16

Pull Request - State: closed - Opened by zqevans 4 months ago

#69 - Add diffusion documentation

Pull Request - State: closed - Opened by zqevans 4 months ago

#68 - fix ValueError: high is out of bounds for int32

Pull Request - State: closed - Opened by underdogest 4 months ago

#67 - CUDA not recognized, also -1 seed not working.

Issue - State: open - Opened by Atoli 4 months ago - 4 comments

#66 - `ImportError: cannot import name 'packaging' from 'pkg_resources'` from `run_gradio.py`

Issue - State: open - Opened by YoungPhlo 4 months ago - 3 comments

#65 - Possibility of the generate songs like suno

Issue - State: open - Opened by Xiaoyiyong555 4 months ago - 25 comments

#64 - Fixing bugs in audio saving of AutoEncoder

Pull Request - State: closed - Opened by yukara-ikemiya 5 months ago - 1 comment

#63 - Loss about vae and diffusion

Issue - State: open - Opened by 980202006 5 months ago - 14 comments

#62 - Is it neccesarry that the latent space has approximately unit variance？

Issue - State: closed - Opened by hongwen-sun 5 months ago - 2 comments