Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

RuntimeError: Backend "sox_io" is not one of available backends: ['soundfile']. #93

Merged
merged 2 commits into from
Jul 15, 2021
Merged

RuntimeError: Backend "sox_io" is not one of available backends: ['soundfile']. #93

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d4b6be0
RuntimeError: Backend "sox_io" is not one of available backends: ['so…
faroit Jul 15, 2021
4861ee5
update socs
faroit Jul 15, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion docs/inference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ The separation can be controlled with additional parameters that influence the p
| `--residual` | computes a residual target, for custom separation scenarios when not all targets are available (at the expense of slightly less performance). E.g vocal/accompaniment can be performed with `--targets vocals --residual`. | not set |
| `--softmask` | if activated, then the initial estimates for the sources will be obtained through a ratio mask of the mixture STFT, and not by using the default behavior of reconstructing waveforms by using the mixture phase. | not set |
| `--wiener-win-len <int>` | Number of frames on which to apply filtering independently | `300` |
| `--audio-backend <str>` | choose audio loading backend, either `sox_io`, `soundfile` or `stempeg` (which needs additional installation requirements) | `sox_io` |
| `--audio-backend <str>` | choose audio loading backend, either `sox_io`, `soundfile` or `stempeg` (which needs additional installation requirements) | [torchaudio default](https://pytorch.org/audio/stable/backend.html) |
| `--aggregate <str>` | if provided, must be a string containing a valid expression for a dictionary, with keys as output target names, and values a list of targets that are used to build it. For instance: `{ "vocals": ["vocals"], "accompaniment": ["drums", "bass", "other"]}` | `None` |
| `--filterbank <str>` | filterbank implementation method. Supported: `['torch', 'asteroid']`. While `torch` is ~30% faster compared to `asteroid` on large FFT sizes such as 4096, asteroids STFT maybe be easier to be exported for deployment. | `torch` |

Expand Down
7 changes: 3 additions & 4 deletions openunmix/cli.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,9 +64,8 @@ def separate():
parser.add_argument(
"--audio-backend",
type=str,
default="sox_io",
help="Set torchaudio backend "
"(`sox_io`, `sox`, `soundfile` or `stempeg`), defaults to `sox_io`",
help="Sets audio backend. Default to torchaudio's default backend: See https://pytorch.org/audio/stable/backend.html"
"(`sox_io`, `sox`, `soundfile` or `stempeg`)",
)

parser.add_argument(
Expand Down Expand Up @@ -114,7 +113,7 @@ def separate():
)
args = parser.parse_args()

if args.audio_backend != "stempeg":
if args.audio_backend != "stempeg" and args.audio_backend is not None:
torchaudio.set_audio_backend(args.audio_backend)

use_cuda = not args.no_cuda and torch.cuda.is_available()
Expand Down