format

README.md

@@ -11,7 +11,7 @@
 
 ## Features
 
- See [feature list](doc/src/feature_list.md) for more information. 
+ See [feature list](doc/src/feature_list.md) for more information.
 
 ## Setup
 

deepspeech/frontend/normalizer.py

@@ -179,7 +179,8 @@ class FeatureNormalizer(object):
                 wav_number += batch_size
 
                 if wav_number % 1000 == 0:
-                    logger.info(f'process {wav_number} wavs,{all_number} frames.')
+                    logger.info(
+                        f'process {wav_number} wavs,{all_number} frames.')
 
         self.cmvn_info = {
             'mean_stat': list(all_mean_stat.tolist()),

doc/src/asr_text_backend.md

@@ -98,4 +98,4 @@
 
 ## Text Filter
 
-* 敏感词（黄暴、涉政、违法违禁等）
+* 敏感词（黄暴、涉政、违法违禁等）

doc/src/benchmark.md

@@ -14,4 +14,3 @@ We compare the training time with 1, 2, 4, 8 Tesla V100 GPUs (with a subset of L
 | 8         | 6.95 X |
 
 `utils/profile.sh` provides such a demo profiling tool, you can change it as need.
-

doc/src/chinese_syllable.md

@@ -48,4 +48,4 @@
 ## Zhuyin
 
 * [Bopomofo](https://en.wikipedia.org/wiki/Bopomofo)
-* [Zhuyin table](https://en.wikipedia.org/wiki/Zhuyin_table)
+* [Zhuyin table](https://en.wikipedia.org/wiki/Zhuyin_table)

doc/src/dataset.md

@@ -18,4 +18,4 @@
 
 ### ASR Noise
 
-* [asr-noises](https://github.com/speechio/asr-noises)
+* [asr-noises](https://github.com/speechio/asr-noises)

doc/src/feature_list.md

@@ -58,4 +58,4 @@
 ### Grapheme To Phoneme
 
 * syallable
-* phoneme
+* phoneme

doc/src/ngram_lm.md

@@ -83,4 +83,4 @@ Please notice that the released language models only contain Chinese simplified
 
    ```
    build/bin/build_binary ./result/people2014corpus_words.arps ./result/people2014corpus_words.klm
-   ```
+   ```

doc/src/praat_textgrid.md

@@ -76,7 +76,7 @@ pip3 install textgrid
    tg.read('file.TextGrid')  # 'file.TextGrid' 是文件名
    ```
 
-   tg.tiers属性: 
+   tg.tiers属性:
    会把文件中的所有item打印出来, print(tg.tiers) 的结果如下:
 
    ```text
@@ -86,7 +86,7 @@ pip3 install textgrid
            Interval(1361.89250, 1362.01250, R),
            Interval(1362.01250, 1362.13250, AY1),
            Interval(1362.13250, 1362.16250, T),
-   
+
    ...
            ]
        )
@@ -113,7 +113,7 @@ pip3 install textgrid
    Interval  可以理解为时长
    ```
 
-   
+
 
 2. textgrid库中的对象
    **IntervalTier** 对象:
@@ -148,7 +148,7 @@ pip3 install textgrid
    strict  -- > 返回bool值, 表示是否严格TextGrid格式
    ```
 
-   ​      
+   ​  
 
    **PointTier** 对象:
    方法
@@ -174,7 +174,7 @@ pip3 install textgrid
    name    返回name
    ```
 
-   
+
 
    **Point** 对象:
        支持比较大小, 支持加减运算
@@ -185,7 +185,7 @@ pip3 install textgrid
    time:
    ```
 
-   ​       
+   ​  
 
    **Interval** 对象:
        支持比较大小, 支持加减运算
@@ -250,10 +250,9 @@ pip3 install textgrid
    grids:  --> 返回读取的grids的列表
    ```
 
-   
+
 
 ## Reference
 
 * https://zh.wikipedia.org/wiki/Praat%E8%AF%AD%E9%9F%B3%E5%AD%A6%E8%BD%AF%E4%BB%B6
 * https://blog.csdn.net/duxin_csdn/article/details/88966295
-

doc/src/tools.md

@@ -1,4 +1,3 @@
 # Useful Tools
 
 * [正则可视化和常用正则表达式](https://wangwl.net/static/projects/visualRegex/#)
-

doc/src/tts_text_front_end.md

@@ -23,7 +23,7 @@ Therefore, procedures like stemming and lemmatization are not useful for Chinese
 
 ### Tokenization
 
-**Tokenizing breaks up text data into shorter pre-set strings**, which help build context and meaning for the machine learning model.   
+**Tokenizing breaks up text data into shorter pre-set strings**, which help build context and meaning for the machine learning model.  
 
 These “tags” label the part of speech. There are 24 part of speech tags and 4 proper name category labels in the `**jieba**` package’s existing dictionary.
 
@@ -31,7 +31,7 @@ These “tags” label the part of speech. There are 24 part of speech tags and
 
 ### Stop Words
 
-In NLP, **stop words are “meaningless” words** that make the data too noisy or ambiguous. 
+In NLP, **stop words are “meaningless” words** that make the data too noisy or ambiguous.
 
 Instead of manually removing them, you could import the `**stopwordsiso**` package for a full list of Chinese stop words. More information can be found [here](https://pypi.org/project/stopwordsiso/). And with this, we can easily create code to filter out any stop words in large text data.
 
@@ -188,4 +188,4 @@ TN: 基于规则的方法
 ## Reference
 * [Text Front End](https://slyne.github.io/%E5%85%AC%E5%BC%80%E8%AF%BE/2020/10/03/TTS1/)
 * [Chinese Natural Language (Pre)processing: An Introduction](https://towardsdatascience.com/chinese-natural-language-pre-processing-an-introduction-995d16c2705f)
-* [Beginner’s Guide to Sentiment Analysis for Simplified Chinese using SnowNLP](https://towardsdatascience.com/beginners-guide-to-sentiment-analysis-for-simplified-chinese-using-snownlp-ce88a8407efb)
+* [Beginner’s Guide to Sentiment Analysis for Simplified Chinese using SnowNLP](https://towardsdatascience.com/beginners-guide-to-sentiment-analysis-for-simplified-chinese-using-snownlp-ce88a8407efb)

requirements.txt

@@ -1,5 +1,6 @@
 coverage
 pre-commit
+pybind11
 resampy==0.2.2
 scipy==1.2.1
 sentencepiece
@@ -7,7 +8,6 @@ snakeviz
 SoundFile==0.9.0.post1
 sox
 tensorboardX
+textgrid
 typeguard
 yacs
-pybind11
-textgrid