エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 TacotronとWaveNetのチュートリアル (Part 2)

Tacotron c. Tacotron2 4. Our work: Japanese Tacotron 5. Implementation 3 4. TTS architecture: traditional pipeline - Typical pipeline architecture for statistical parametric speech synthesis - Consists of task-specific models - linguistic model - alignment (duration) model - acoustic model - vocoder 4 5. TTS architecture: End-to-end model - End-to-end model directly converts text to waveform - End-to-end model does not require intermediate feature extraction - Pipeline models accumulate errors across predicted features - End-to-end model's internal blocks are jointly optimized 5 6. [1] Wei Ping, Kainan Peng, Jitong Chen: ClariNet: Parallel Wave Generation in End-to-End Text-to-Speech. End-to-end model: Decoding with attention mechanism 10 Time step 2. - The previous output is fed back to next time step - Assign alignment probabilities to encoded inputs 11.