Batch course narration: from lesson script to chapter audio

1. Install and check your computer

Download the matching installer. On the activation screen, run the compatibility checks and fixed test samples to judge quality and speed on your own computer. Fixed tests cannot use editable text or export finished work. Custom narration requires a paid license and activation. Email freeteamlite@gmail.com with purchase or installation questions. Runtime components and voice packages need an initial internet download; narration can run offline after preparation.

2. Import your script with clear chapters and paragraphs

Create a desktop voice project and import TXT, CSV, TSV, DOCX or PPTX. DOCX extracts body text and tables. For each PPTX slide, the app reads speaker notes first and falls back to slide text if notes are absent. It does not recognize text in images or import PDF files.

Write formulas, abbreviations and numbered items the way you want them spoken. For example, replace ‘length × width’ with ‘length times width’ in the narration script. Give each chapter a clear title. Check the imported order, pronunciations and missing content before generating.

3. Choose an instructor voice before generating the course

Listen to the existing samples and select a voice that suits your lesson. There are 42 voice choices: nine fixed voices and 33 voice-design presets. Character presets need the relevant packages. They lock to a chosen sample by default, then reuse it for later lines. Redefining the voice or using free-form design can change its sound. Website samples do not guarantee identical output. The existing samples keep their original languages; Ryan and Aiden are the fixed English voices.

You can also use 3–30 seconds of clear speech you have permission to use, with an accurate transcript. In the current release, locked and reference voices retain their sample’s expression. Standard preset voices support line-level expression; light mode supports natural delivery only. Generate a short passage of your actual lesson first and check terminology, pauses and pacing.

4. Generate in segments and redo only what changes

Projects can store up to 10,000 lines and 500,000 characters, organized by chapter and role. Long lines are split into shorter segments and processed sequentially. These are project storage limits, not a promise to infer the whole script simultaneously. Speed depends on your computer and mode.

Quick narration and CLI batches support up to 5,000 characters and 100 segments. Each CLI segment is limited to 600 characters and must be split in advance. Work through a long course chapter by chapter. Listen to each result, regenerate segments with script changes or pronunciation issues and compare previous takes before choosing one.

5. Edit and export audio with subtitles

Trim waveforms, adjust volume and add fades in the desktop project while preserving the originals. If needed, import local background music, loop it and lower it during speech. Preview the first 30 seconds of the mix, then export merged WAV, MP3, SRT and ZIP. Make sure you have permission to use the music.

A finished export supports at most 5,000 audio segments and six hours. Export longer courses by chapter. SRT timing follows generated audio segments; it is not word-level recognition and does not automatically align to slide animation or an existing video. Check picture, slide changes and subtitle timing in your video editor.

Check the course before delivery

• Listen to names, numbers, formulas, foreign words and segment boundaries.
• Keep the same instructor sample and check that retakes do not unexpectedly change voices.
• Review chapter volume and background music so the lesson remains clear.
• Check subtitles, picture and pauses in the final course or video before delivery.