Issues
- 3
Torch 2.3 breaks DDP?
#2527 opened by TParcollet - 2
Training regression for Conformer-Transducer models
#2533 opened by asumagic - 0
Huggingface-Aishell get wrong prediction
#2549 opened by Ht-zhang-xianyu - 1
Task ASR Reported: Caught ZeroDivisionError in DataLoader worker process 0.
#2547 opened by sunsdy2018 - 0
- 0
Potential `SpectrogramDrop` bugs
#2542 opened by asumagic - 8
Typing syntax not supported in 3.7/3.8
#2536 opened by hughhop - 2
Tacotron2.decoder.infer behaves incorrectly
#2511 opened by QuentinJGMace - 10
Training twice as long with Torch > 1.11
#2531 opened by Craya - 1
Math Domain Error in Pretraining tutorial.
#2535 opened by thedragoncoder - 2
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!
#2516 opened by shripadk - 6
Adapters + LLama -- re-design.
#2526 opened by TParcollet - 1
speechbrain inference classifier error
#2457 opened by JGanson - 7
not able to import 'HuggingFaceWhisper' from speechbrain.lobes.models.huggingface_whisper
#2522 opened by aquorio15 - 2
spkrec-ecapa-voxceleb-mel-spec model modifies mel spectrum in place when used with CPU
#2390 opened by jacobjwebber - 9
Can't reproduce pretraining results for Wav2vec2 using LibriSpeech recipe
#2512 opened by GasserElbanna - 9
- 4
Issues regarding discrete WavLM and discrete HuBERT
#2453 opened by anupsingh15 - 2
- 6
- 2
Circular import in ESC-50 classification recipe
#2501 opened by lucadellalib - 8
Circular Import Error
#2500 opened by Drew-Wagner - 1
Hi! LibriSpeech char training!!
#2385 opened by scj0709 - 1
- 8
Incorrect transformer mask size
#2344 opened by egaznep - 3
Not Able to use 'speechbrain/spkrec-ecapa-voxceleb' embeddings in Google Colab.
#2438 opened by itsdivgithub - 0
- 0
PyPI install incorrectly ships a `tests` package
#2491 opened by asumagic - 1
`speechbrain/__init__.py` is ignored in pip editable installs for scripts out of repo
#2484 opened by asumagic - 1
Inference interface uncompatible with Python < 3.9
#2452 opened by lucadellalib - 4
- 0
[Feature Request]: AdaMER-CTC for ASR task training
#2468 opened by tomschelsen - 1
Unable to use model trained from enhancement template
#2422 opened by Zahuczky - 0
- 3
- 2
- 3
AMP at inference time
#2415 opened by lucadellalib - 2
Wav2Vec2Pretrain (HFTransformersInterface implementation) samples padded values for mask_time_indices and negative_sample_indices
#2386 opened by porfirythelaw - 5
Train loop may crash during checkpointing
#2397 opened by kokamido - 7
Possible NCCL-level deadlock during checkpointing
#2401 opened by kokamido - 0
Error during encoding
#2408 opened by husnejahan - 7
RuntimeError when processing VAD on short audio
#2334 opened by gaspardpetit - 1
[Feature Request]: Load a speechbrain-fine-tuned huggingface model checkpoint with the huggingface interface
#2342 opened by MaloMn - 0
[Feature Request]: Improve DAC interface
#2339 opened by lucadellalib - 1
[Feature Request]: Stree info in G2P output
#2337 opened by ease-zh - 9
- 0
loading fully trained brain (a brain who finished training) and then evaluating it with brain.evaluate. will cause a crash when using hpopt context but not reporting in test stage
#2358 opened by avishaiElmakies - 0
DAC cannot be imported
#2332 opened by lucadellalib - 1
VAD Example Code Broken: Need hparams['sample_rate']
#2322 opened by weilun-enso - 2
No module named 'speechbrain.inference'
#2325 opened by chrisspen