Submit the complete script
Paste your full narration in single-script mode or upload a UTF-8 text file. The server splits the text according to the speech engine's input limits and assembles the resulting audio. You do not have to send a separate request for every synthesis chunk.
From script to audio
- Prepare the text
Keep paragraphs and punctuation where you want natural pauses. Check the current studio limits before submitting.
- Choose the voice and format
Choose a first-release voice for Mandarin Chinese or English, adjust speed, and select WAV or MP3. The first public release uses ordinary narration without style prompts.
- Follow the task and download
Check progress using the same task. Listen to the completed result, then download it. If a task fails, resolve the cause before resuming unfinished work.
Long-form input or a batch?
- Long-form input
- A complete script is submitted as one item. Narrivela manages its internal synthesis chunks.
- Batch input
- Each script remains an independent item with its own audio and place in the order. See batch generation.
- Output and limits
- WAV and MP3 are supported. The studio shows available voices and current input and queue limits. There is no promise of unlimited text or concurrency.
Try the workflow
Open the studio · Listen to the sample · Read the parameter guide