speechbrain · Issues· 188 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #3071
Training efficiency regression since SpeechBrain 1.1.0
bugUpdated Sep 8, 2026 - #3076
Verify evals on Papers with Code
Updated Sep 4, 2026 - #3074
FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
bugUpdated Aug 26, 2026 - #3070
make_pe() in class RelPosEncXL
bugUpdated Aug 9, 2026 - #3058
Feature request: Add FunASR/SenseVoice model integration
Updated May 31, 2026 - #2841
SpeechBrain should rely on steps and epochs
bugUpdated Apr 17, 2026 - #3041
HyperConformer and Conformer Achieve Higher WER on test-other When Trained on 100h LibriSpeech
bugUpdated Mar 14, 2026 - #3012
AttributeError: module 'torchaudio' has no attribute 'list_audio_backends'
bugUpdated Feb 16, 2026 - #2663
SpeechLLM and Whisper
enhancementUpdated Feb 7, 2026 - #2847
Add PC-GITA Recipe with VocalFeatures
recipesUpdated Feb 7, 2026 - #2661
SpeechBrain and ONNX
bugenhancementUpdated Feb 7, 2026 - #2660
SpeechBrain pipelines
enhancementUpdated Feb 7, 2026 - #3014
I can not download RIR.zip and noises.zip for voxceleb train speaker rec
bugUpdated Dec 11, 2025 - #2710
Harmonize Adapters and MultiWOZ w/ QLoRA
Updated Nov 24, 2025 - #2995
`LazyModule` not playing well with `%autoreload`
bugUpdated Nov 20, 2025