audio

audio copied to clipboard

Add Stereo to Mono Convertions

18

## 🚀 Feature Make mono to stereo or stereo to mono conversion ## Motivation You Guys have made an amazing job, but stereo to mono and vice versa is simple,...

RemonComputer

Add stereo to mono transform

5

Implements a transformation for converting multi-channel audio to mono audio. The conversion is done by just taking the average of each channel and then dividing it by the number of...

jjmmchema

About HUBERT_ASR_BASE pipeline

3

### 🚀 The feature Hello! I want to use the finetuned HuBERT_base model. However, in torchaudio.pipelines, there has only HUBERT_ASR_LARGE and HUBERT_ASR_XLARGE. What should I do to get a HUBERT_ASR_BASE...

LYPinASR

Add a torch implementation of "convolution reverb"

5

### 🚀 The feature A pure Pytorch implementation of the "convolution reverb" like described in https://pytorch.org/audio/stable/tutorials/audio_data_augmentation_tutorial.html#simulating-room-reverberation This should be implemented like "pitch shift" both in "functional" and as a module....

gwenzek

Hardcoded padding mode in functional.spectrogram

3

I wonder why the padding mode is hardcoded in functional.spectrogram? Maybe add a parameter to support reflect padding? https://github.com/pytorch/audio/blob/main/torchaudio/functional/functional.py#L117

gormat

### 🐛 Describe the bug The Kaldi feature extraction algorithm returns two tensors, `pitch` and `NCCF`: ```python import torch import torchaudio from torchaudio.utils import download_asset import torchaudio.functional as F SAMPLE_SPEECH...

ziadloo

Guidance on docs versioning

1

I've received feedback that the versioning UI is both not obvious enough and also that it is not clear what it means. Adding some copy to the page to clarify.

carljparker

torchaudio.transforms.SpectralCentroid supports only real valued input

4

### 🐛 Describe the bug The SpectralCentroid transform supports only real valued inputs. If complex values are provided, it yields an error : ```RuntimeError: Cannot have onesided output if window...

rgt-yncrea

Wav2vec2.0 Pretrained model gives different emission results for different batch size input.

12

### 🐛 Describe the bug I am trying to modify the example/asr/librispeech_ctc_decode/inference.py to a batch mode. Here is my script: https://gist.github.com/yuekaizhang/f20904cfaf23e457a744f08ea19ce18e#file-inference_bug-py-L55 However, I found that with different batch_size, the WER...

yuekaizhang

Update stable symlink to 2.3.0.

1

Update stable symlink to 2.3.0.

ahmadsharif1