Update README.md

README.md

@@ -6,12 +6,18 @@ tags:
 - jvs
 - pyopenjtalk
 - speech-to-text
+pipeline_tag: text-to-speech
 ---
 
 # SpeechT5 (TTS task) for Japanese
-SpeechT5 model fine-tuned for speech synthesis (text-to-speech) on [JVS]("https://sites.google.com/site/shinnosuketakamichi/research-topics/jvs_corpus").
+SpeechT5 model fine-tuned for Japanese speech synthesis (text-to-speech) on [JVS]("https://sites.google.com/site/shinnosuketakamichi/research-topics/jvs_corpus").
+This model utilizes the JVS dataset which encompasses 100 speakers.
+From this dataset, speaker embeddings were crafted, segregating them based on male and female voice types, and producing a unique speaker embedding vector.
+This 16-dimensional speaker embedding vector is designed with an aim to provide a voice quality that is independent of any specific speaker.
+
 Trained from [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts).
-Modified tokenizer powered by [Open Jtalk](https://open-jtalk.sp.nitech.ac.jp/)
+Modified tokenizer powered by [Open Jtalk](https://open-jtalk.sp.nitech.ac.jp/).
+
 
 # Model description
 See [original model card](https://huggingface.co/microsoft/speecht5_tts#model-description)