New journal paper (pre-proof version):
New journal paper:
New preprint:
- On Training Targets for Deep Learning Approaches to Clean Speech Magnitude Spectrum Estimation
- The code for this study can be found at: https://github.com/anicolson/tgt_20.
- Introduction
- How does Deep Xi work?
- Current networks
- Available models
- Results
- DeepMMSE
- Installation
- How to use Deep Xi
- Current issues and potential areas of improvement
- Where can I get a dataset for Deep Xi?
- Which audio do I use with Deep Xi?
- Citation guide
Deep Xi is implemented in TensorFlow 2 and can be used for speech enhancement, noise estimation, mask estimation, and as a front-end for robust ASR. Deep Xi (where the Greek letter 'xi' or ξ is pronounced /zaɪ/ and is the symbol used in the literature for the a priori SNR) is a deep learning approach to a priori SNR estimation that was proposed in [1] and is implemented in TensorFlow 2. Some of its use cases include:
- It can be used by minimum mean-square error (MMSE) approaches to speech enhancement like the MMSE short-time spectral amplitude (MMSE-STSA) estimator.
- It can be used by MMSE-based noise PSD estimators, as in DeepMMSE [2].
- Estimate the ideal binary mask (IBM) for missing feature approaches or the ideal ratio mask (IRM).
- A front-end for robust ASR
A training example is shown in Figure 2. A deep neural network (DNN) within the Deep Xi framework is fed the noisy-speech short-time magnitude spectrum as input. The training target of the DNN is a mapped version of the instantaneous a priori SNR (i.e. mapped a priori SNR). The instantaneous a priori SNR is mapped to the interval [0,1]
to improve the rate of convergence of the used stochastic gradient descent algorithm. The map is the cumulative distribution function (CDF) of the instantaneous a priori SNR, as given by Equation (13) in [1]. The statistics for the CDF are computed over a sample of the training set. An example of the mean and standard deviation of the sample for each frequency bin is shown in Figure 3. The training examples in each mini-batch are padded to the longest sequence length in the mini-batch. The sequence mask is used by TensorFlow to ensure that the DNN is not trained on the padding. During inference, the a priori SNR estimate is computed from the mapped a priori SNR using the sample statistics and Equation (12) from [2].
![]() |
---|
Figure 2: A training example for Deep Xi. Generated using |
Recurrent neural networks (RNNs) and temporal convolutional networks (TCNs), are available:
- ResLSTM & ResBiLSTM: Residual long short-term memory (LSTM) network and residual bidirectional LSTM (ResBiLSTM) network [1].
- ResNet: Residual network [2].
- RDLNet: Residual-dense lattice network [3].
Deep Xi utilising a ResNet TCN (Deep Xi-ResNet) was proposed in [2]. It uses bottleneck residual blocks and a cyclic dilation rate. The network comprises of approximately 2 million parameters and has a contextual field of approximately 8 seconds. An example of Deep Xi-ResNet is shown in Figure 4. Deep Xi utilising a ResLSTM network (Deep Xi-ResLSTM) was proposed in [1]. Each of its residual blocks contain a single LSTM cell. The network comprises of approximately 10 million parameters.
There are multiple Deep Xi versions, comprising of different networks and restrictions. An example of the ver
naming convention is resnet-1.0c
. The network type is given at the start of ver
. Versions with c are causal. Versions with n are non-causal. The version iteration is also given, i.e. 1.0
. Here are the current versions:
resnet-1.1n
(available in the model
directory)
resnet-1.1c
(available in the model
directory)
resbilstm-1.0n
(available soon)
reslstm-1.0c
(available soon)
Each available model is trained using the Deep Xi dataset (seperate models are trained on the DEMAND-Voice bank training set for the DEMAND-Voice bank test set results below). Please see run.sh
for more details about these networks.
DEMAND -- Voice Bank test set
Objective scores obtained on the DEMAND--Voicebank test set described here. Each Deep Xi model is trained on the DEMAND--Voicebank training set. As in previous works, the objective scores are averaged over all tested conditions. CSIG, CBAK, and COVL are mean opinion score (MOS) predictors of the signal distortion, background-noise intrusiveness, and overall signal quality, respectively. PESQ is the perceptual evaluation of speech quality measure. STOI is the short-time objective intelligibility measure (in %). The highest scores attained for each measure are indicated in boldface.
Method | Gain | Causal | CSIG | CBAK | COVL | PESQ | STOI | SegSNR |
---|---|---|---|---|---|---|---|---|
Noisy speech | -- | -- | 3.35 | 2.44 | 2.63 | 1.97 | 92 (91.5) | -- |
Wiener | Yes | 3.23 | 2.68 | 2.67 | 2.22 | -- | -- | |
SEGAN | -- | No | 3.48 | 2.94 | 2.80 | 2.16 | 93 | |
WaveNet | -- | No | 3.62 | 3.23 | 2.98 | -- | -- | |
MMSE-GAN | -- | No | 3.80 | 3.12 | 3.14 | 2.53 | 93 | -- |
Deep Feature Loss | -- | Yes | 3.86 | 3.33 | 3.22 | -- | -- | -- |
Metric-GAN | -- | No | 3.99 | 3.18 | 3.42 | 2.86 | -- | -- |
Deep Xi-ResLSTM (1.0c) | MMSE-LSA | Yes | 4.01 | 3.25 | 3.34 | 2.65 | 91 (91.0) | 8.2 |
Deep Xi-ResNet (1.0c) | MMSE-LSA | Yes | 4.14 | 3.32 | 3.46 | 2.77 | 93 (93.2) | -- |
Deep Xi-ResNet (1.0n) | MMSE-LSA | No | 4.28 | 3.46 | 3.64 | 2.95 | 94 (93.6) | -- |
Deep Xi-ResNet (1.1c) | MMSE-LSA | Yes | 4.24 | 3.40 | 3.59 | 2.91 | 94 (93.5) | 8.4 |
Deep Xi-ResNet (1.1n) | MMSE-LSA | No | 4.35 | 3.52 | 3.71 | 3.03 | 94 (94.1) | 9.3 |
Deep Xi Test Set
Average objective scores obtained over the conditions in the test set of the Deep Xi dataset. Each Deep Xi model is trained on the test set of the Deep Xi dataet. SNR levels between -10 dB and 20 dB are considered only. MOS-LQO is the mean opinion score (MOS) objective listening quality score obtained using Wideband PESQ. PESQ is the perceptual evaluation of speech quality measure. STOI is the short-time objective intelligibility measure (in %). eSTOI is extended STOI. Results for each condition can be found in log/results
Method | Gain | Causal | MOS-LQO | PESQ | STOI | eSTOI |
---|---|---|---|---|---|---|
Deep Xi-ResNet (1.0c) | MMSE-STSA | Yes | 1.90 | 2.34 | 80.92 | 65.90 |
Deep Xi-ResNet (1.0c) | MMSE-LSA | Yes | 1.92 | 2.37 | 80.79 | 65.77 |
Deep Xi-ResNet (1.0c) | SRWF | Yes | 1.87 | 2.31 | 80.98 | 65.94 |
Deep Xi-ResNet (1.0c) | cWF | Yes | 1.92 | 2.34 | 81.11 | 65.79 |
Deep Xi-ResNet (1.0c) | WF | Yes | 1.75 | 2.21 | 78.30 | 63.96 |
Deep Xi-ResNet (1.0c) | IBM | Yes | 1.38 | 1.73 | 70.85 | 55.95 |
Deep Xi-ResNet (1.1c) | MMSE-LSA | Yes | 1.92 | 2.36 | 81.88 | 66.44 |
Deep Xi-ResNet (1.1n) | MMSE-LSA | No | 2.02 | 2.48 | 83.90 | 69.50 |
DeepMMSE: A Deep Learning Approach to MMSE-Based Noise Power Spectral Density Estimation.
To save noise PSD estimate .mat
files from DeepMMSE, please use the following:
./run.sh VER="resnet-1.1n" INFER=1 GAIN="deepmmse"
Prerequisites for GPU usage:
To install:
git clone https://github.com/anicolson/DeepXi.git
virtualenv --system-site-packages -p python3 ~/venv/DeepXi
source ~/venv/DeepXi/bin/activate
cd DeepXi
pip install -r requirements.txt
Use run.sh
to configure and run Deep Xi.
Inference: To perform inference and save the outputs, use the following:
./run.sh VER="resnet-1.1n" INFER=1 GAIN="mmse-lsa"
Please look in thoth/args.py
for available gain functions and run.sh
for further options.
Testing: To perform testing and get objective scores, use the following:
./run.sh VER="resnet-1.1n" TEST=1 GAIN="mmse-lsa"
Please look in log/results
for the results.
Training:
./run.sh VER="resnet-1.1n" TRAIN=1
Ensure to delete the data directory before training. This will allow training lists and statistics for your training set to be saved and used. To retrain from a certain epoch, set --resume_epoch
in run.sh
to the desired epoch.
If you would like to contribute to Deep Xi, please investigate the following and compare it to current models:
- Currently, the ResLSTM network is not performing as well as expected (when compared to TensorFlow 1.x performance).
Open-source training and testing sets are available for Deep Xi on IEEE DataPort:
[4] Deep Xi dataset (training, validation, and test set): http://dx.doi.org/10.21227/3adt-pb04.
[5] Test set from the original Deep Xi paper: http://dx.doi.org/10.21227/0ppr-yy46.
The MATLAB scripts used to generate these sets can be found in set
.
Deep Xi operates on mono/single-channel audio (not stereo/dual-channel audio). Single-channel audio is used due to most cell phones using a single microphone. The available trained models operate on a sampling frequency of f_s=16000
Hz, which is currently the standard sampling frequency used in the speech enhancement community. The sampling frequency can be changed in run.sh
. Deep Xi can be trained using a higher sampling frequency (e.g. f_s=44100
Hz), but this is unnecessary as human speech rarely exceeds 8 kHz (the Nyquist frequency of f_s=16000
Hz is 8 kHz). The available trained models operate on a window duration and shift of T_d=32
ms and T_s=16
ms, respectively. To train a model on a different window duration and shift, T_d
and T_s
can be changed in run.sh
. Currently, Deep Xi supports .wav
, .mp3
, and .flac
audio codecs. The audio codec and bit rate does not affect the performance of Deep Xi.
The following is already configured in the Deep Xi dataset.
Training set
The filenames of the waveforms in the train_clean_speech
and train_noise
directories are not restricted. There can be a different number of waveforms in each. The Deep Xi framework utilises each of the waveforms in train_clean_speech
once during an epoch. For each train_clean_speech
waveform of a mini-batch, the Deep Xi framework selects a random section of a randomely selected waveform from train_noise
(that is at a length greater than or equal to the train_clean_speech
waveform) and adds it to the train_clean_speech
waveform at a randomly selected SNR level (the SNR level range can be set in run.sh
).
Validation set
As the validation set must not change from epoch to epoch, a set of restrictions apply to the waveforms in val_clean_speech
and val_noise
. There must be the same amount of waveforms in val_clean_speech
and val_noise
. One waveform in val_clean_speech
corresponds to only one waveform in val_noise
, i.e. a clean speech and noise validation waveform pair. Each clean speech and noise validation waveform pair must have identical filenames and and an identical number of samples. Each clean speech and noise validation waveform pair must have the SNR level (dB) that they are to be mixed at placed at the end of their filenames. The convention used is _XdB
, where X
is replaced with the desired SNR level. E.g. val_clean_speech/NAME_-5dB.wav
and val_noise/NAME_-5dB.wav
. An example of the filenames for a clean speech and noise validation waveform pair is as follows: val_clean_speech/198_19-198-0003_Machinery17_15dB.wav
and val_noise/198_19-198-0003_Machinery17_15dB.wav
.
Test set
The filenames of the waveforms in the test_noisy_speech
directory are not restricted. This is all that is required if you want inference outputs from Deep Xi, i.e. ./run.sh VER="ANY_NAME" INFER=1
. If you are obtaining objective scores by using ./run.sh VER="ANY_NAME" TEST=1
, then reference waveforms for the objective measures need to be placed in test_clean_speech
. The waveforms in test_clean_speech
and test_noisy_speech
that correspond to each other must have the same number of samples (i.e. the same sequence length). The filename of the waveform in test_clean_speech
that corresponds to a waveform in test_noisy_speech
must be contained in the corresponding test noisy speech waveforn filename. E.g. if the filename of a test noisy speech waveform is test_noisy_speech/61-70968-0000_SIGNAL021_-5dB.wav
, then the filename of the corresponding test clean speech waveform must be contained in the filename of the test noisy speech waveform: test_clean_speech/61-70968-0000.wav
. This is because a test clean speech waveform may be used as a reference for multiple waveforms in test_noisy_speech
(e.g. test_noisy_speech/61-70968-0000_SIGNAL021_0dB.wav
, test_noisy_speech/61-70968-0000_SIGNAL021_5dB.wav
, and test_noisy_speech/61-70968-0000_SIGNAL021_10dB.wav
are additional test noisy speech waveforms that the test clean speech waveform from the previous example is a reference for).
Please cite the following depending on what you are using:
- The Deep Xi framework is proposed in [1].
- If using Deep Xi-ResLSTM, please cite [1].
- If using Deep Xi-ResNet, please cite [1] and [2].
- If using DeepMMSE, please cite [2].
- If using Deep Xi-RDLNet, please cite [1] and [3].
- If using Deep Xi dataset, please cite [4].
- If using the Test Set From 10.1016/j.specom.2019.06.002, please cite [5].
[4] Aaron Nicolson, "Deep Xi dataset", IEEE Dataport, 2020. [Online]. Available: http://dx.doi.org/10.21227/3adt-pb04.
[5] Aaron Nicolson, "Test Set From 10.1016/j.specom.2019.06.002", IEEE Dataport, 2020. [Online]. Available: http://dx.doi.org/10.21227/0ppr-yy46.