原创 Chapter6_Vocoder
文章目錄1 Introduction2 WaveNet2.1 WaveNet的架構2.2 Softmax Distribution2.3 Causal Convolution和Dilated Convolution2.4 Gate
2020-07-05 05:24:14
19
原创 Chapter4-1_Speech_Synthesis(Tacotron)
文章目錄1 TTS before End-to-end2 Tacotron2.1 Encoder2.2 Attention2.3 Decoder2.4 Post processing3 How good is Tacotron?
2020-07-05 05:24:14
2
原创 Chapter4-2_Speech_Synthesis(More than Tacotron)
文章目錄1 Mispronunciation2 More information for Encoder3 Attention4 Fast Speech and DurIAN5 Dual Learning5 Controllabl
2020-07-05 05:24:14
12
原创 Chapter5_Speaker_Verification
文章目錄1 Task Introduction2 模型架構3 模型介紹3.1 i-vector3.2 d-vector3.3 x-vector3.4 more4 End to End 本文爲李弘毅老師【Speaker Verifi
2020-07-05 05:24:14
4
原创 Chapter3-2_Speech Separation(TasNet)
文章目錄1 TasNet總體架構2 Encoder和Decoder3 Separator4 TasNet回顧5 More5.1 Unknown number of speakers5.2 Multiple microphones5
2020-07-05 05:24:14
20
原创 Chapter3-1_Speech Separation(Deep Clustering, PIT)
文章目錄1 內容簡述2 評價指標2.1 Signal-to-noise ratio (SNR)2.2 Scale invariant signal-to-distortion ratio (SI-SDR)2.3 其他的評價指標3
2020-05-27 05:48:44
6
原创 Chapter2-1_Voice Conversion(Feature Disentangle)
文章目錄1 什麼是Voice Conversion2 實際實現中的細節3 根據數據集分類4 Feature disentangle5 訓練技巧 本文爲李弘毅老師【Voice Conversion - Feature Disenta
2020-05-18 21:28:00
12
原创 Chapter2-2_Voice Conversion(CycleGAN and StarGAN)
文章目錄1 內容簡述2 CycleGAN3 StarGAN 本文爲李弘毅老師【Voice Conversion - CycleGAN and StarGAN】的課程筆記，課程視頻youtube地址，點這裏👈(需翻牆)。 下文中用到
2020-05-18 21:28:00
3
原创 搞懂Transformer
文章目錄1 內容簡述2 seq2seq的常用模塊3 Self-attention4 Multi-head Self-attention5 Positional Encoding6 Transformer 文爲李弘毅老師【Trans
2020-05-06 02:24:44
原创 Chapter1-3_Speech_Recognition(CTC, RNN-T and more)
文章目錄1 CTC2 RNN-T3 Neural Transducer4 Monotonic Chunkwise Attention5 小結 本文爲李弘毅老師【Speech Recognition - CTC, RNN-T and
2020-05-06 02:24:44
原创 Chapter1-7_Speech_Recognition(Language Modeling)
文章目錄1 爲什麼需要Language Model2 N-gram3 Continuous LM3 NN-based LM4 RNN-based LM5 合併LAS和LM5.1 shallow fusion5.2 deep fus
2020-05-06 02:24:44
原创 【課程筆記】李弘毅2020 Deep Learning for Human Language Processing
簡要說明 這是我在學習李弘毅老師的2020春季課程【Deep Learning for Human Language Processing】時做的課程筆記。寫課程筆記的初衷是爲了幫助自己之後快速的回顧複習，因爲我記性不好，不做筆記
2020-05-06 02:24:44
31
原创 Chapter1-2_Speech_Recognition(LAS)
文章目錄1 內容簡述2 模型詳述2.1 ListenRNN Encoder1D-CNN EncoderSelf-attentinon EncoderDown Sampling2.2 Attend2.3 Spell2.4 Beam
2020-05-06 02:24:44
原创 Chapter1-5_Speech_Recognition(Alignment of HMM, CTC and RNN-T)
文章目錄1 爲什麼需要Alignment2 窮舉所有的alignment2.1 HMM的對齊2.2 CTC的對齊2.3 RNN-T的對齊3 小結 本文爲李弘毅老師【Speech Recognition - Alignment of
2020-05-06 02:24:44
1
原创 搞懂RNN
文章目錄1 什麼是RNN2 LSTM3 Training3.1 Learning Target3.2 爲什麼難train4 應用舉例4.1 Many To One4.2 Many To Many4.3 其他 本文爲李弘毅老師【Re
2020-05-06 02:24:44