How can i increase length of the text and length of the audio file?

i was trained M-AI-LABs female EN-UK dataset. i am getting very good results. but it is only detect upto certain limits only.
i was given longer text (3 lines-15 words), that inference result is,

Once upon a time there lived a lion in a forest. One day after a heavy meal. It was sleeping under a tree. After a while, there came a mouse and it started to play on the lion. Suddenly the lion got up with anger and looked for those who disturbed its nice sleep. Then it saw a small mouse standing trembling with fear. The lion jumped on it and started to kill it. The mouse requested the lion to forgive it. The lion felt pity and left it. The mouse ran away.

story_inference.wav.zip (1.2 MB)

but in some text, it is not giving good results.

Our primary focus is to pioneer the application for it and overlapping technologies to protect IP, the security of data and financial transactions and offer cost effective solutions. Rapidly expand our innovative its platform, as well as provide the company a command center for our smtcity technologies.

infer.wav.zip (1.9 MB)

sir what is the problem for this one?

any one give me suggestions.

Thank you.

It is better to split a text into sentences and give them separately to the model. It is a hard problem to convert such a long text without splitting. You might even have a memory overflow since the memory usage is linearly correlated with the length of the text. Hope it helps.

1 Like