文件列表:

3\audiolib.py (2119, 2020-01-11)
3\noisyspeech_synthesizer.cfg (2231, 2020-09-27)
3\noisyspeech_synthesizer.py (5910, 2020-09-16)
3\requirements.txt (88, 2020-06-24)
3 (0, 2020-10-06)

# Microsoft Scalable Noisy Speech Dataset (MS-SNSD) * 此数据集包含大量干净的语音文件和各种环境噪声文件的。wav格式采样在16 kHz。 * 该数据集的主要应用是训练深度神经网络(DNN)模型来抑制背景噪声。但它可以用于其他音频和语音应用。 * 我们提供了在不同信噪比(SNR)条件下混合干净语音和噪声的配方,以生成大的噪声语音数据集。可根据应用要求配置信噪比条件及所需的数据小时数。 * 这个数据集的规模将会继续增长,因为我们鼓励研究人员和从业者通过添加更多清晰的语音和噪声片段来为这个数据集做出贡献。 * 该数据集将极大地帮助学院和行业的研究人员和从业者开发更好的模型。 * 我们还提供了不同于训练集的测试集来评估开发的模型。 * 我们提供html代码,用于构建两个人类智能任务(HIT)众包应用程序,允许用户对噪声音频剪辑进行评级。我们实现了一个绝对类别评级(ACR)应用程序 ## Prerequisites - Python 3.0 and above - pysoundfile (pip install pysoundfile) - ($ pip install -r requirements.txt) ## Please cite us if you use this dataset [@article{reddy2019scalable, title={A Scalable Noisy Speech Dataset and Online Subjective Test Framework$\}$$\}$}, author={Reddy, Chandan KA and Beyrami, Ebrahim and Pool, Jamie and Cutler, Ross and Srinivasan, Sriram and Gehrke, Johannes}, journal={Proc. Interspeech 2019}, pages={1816--1820}, year={2019} }](https://www.isca-speech.org/archive/Interspeech_2019/pdfs/3087.pdf) ## MS-SNSD Dataset # Training and test sets 1. Clean Speech data for training is present in the directory 'CleanSpeech' 2. Noise data for training is present in the directory 'Noise' 3. Noisy Speech for testing is present in the directory 'noisy_test' 4. Clean Speech corresponding to noisy speech test data is present in the directory 'clean_test' Download the data onto your local machine. ## Usage 1. Clone the repo to your local directory 2. Download clean speech and noise datasets into the same directory with scripts 3. The repo contains the following files: - 'audiolib.py' - 'noisyspeech_synthesizer.cfg' - 'noisyspeech_synthesizer.py' - 'requirements.txt' 4. 在配置文件中指定您的需求(noisyspeech_synthesizer.cfg) - 请指定采样率、音频格式、音频长度、静音长度、所需的噪声语音总小时数和所需的信噪比(SNR)水平。 - 指定要排除的噪音文件。例如:noise_types_exclude: Babble, Traffic。“None”表示没有要排除的文件。 - 如果噪音和语音目录与脚本不在同一个目录中,请指定路径。 5. noisy speech和相应的clean speech和noise文件将分别位于“noisyspeech h_training”、“cleanspeech h_training”和“Noise_training”目录中。 6. 为了便于使用,请确保配置文件与(noisyspeech h_synthesizer.py)在同一个目录中。 7. 现在运行(python noisyspeech_synthesizer.py)来生成有噪声的语音片段。 ## Dataset licenses MICROSOFT PROVIDES THE DATASETS ON AN "AS IS" BASIS. MICROSOFT MAKES NO WARRANTIES, EXPRESS OR IMPLIED, GUARANTEES OR CONDITIONS WITH RESPECT TO YOUR USE OF THE DATASETS. TO THE EXTENT PERMITTED UNDER YOUR LOCAL LAW, MICROSOFT DISCLAIMS ALL LIABILITY FOR ANY DAMAGES OR LOSSES, INLCUDING DIRECT, CONSEQUENTIAL, SPECIAL, INDIRECT, INCIDENTAL OR PUNITIVE, RESULTING FROM YOUR USE OF THE DATASETS. The datasets are provided under the original terms that Microsoft received such datasets. See below for more information about each dataset. The datasets used in this project are licensed as follows: 1. Clean speech: * PTDB-TUG: Pitch Tracking Database from Graz University of Technology https://www.spsc.tugraz.at/databases-and-tools/ptdb-tug-pitch-tracking-database-from-graz-university-of-technology.html; License: http://opendatacommons.org/licenses/odbl/1.0/ * Edinburgh 56 speaker dataset: https://datashare.is.ed.ac.uk/handle/10283/2791; License: https://datashare.is.ed.ac.uk/bitstream/handle/10283/2791/license_text?sequence=11&isAllowed=y 2. Noise: * Freesound: https://freesound.org/ Only files with CC0 licenses were selected; License: https://creativecommons.org/publicdomain/zero/1.0/ * Demand: https://zenodo.org/record/1227121#.XRKKxYhKiUk; License: https://creativecommons.org/licenses/by-sa/3.0/deed.en_CA ## Code license MIT License Copyright (c) Microsoft Corporation. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：