Split One Synthesis into Many Audio Segments
12-09-2025 , 26-04-2026
Split one synthesis into many downloadable audio segments using the <cut/> tag. One Convert call produces a clean, separately-named file for every section of your text — no Audacity, no manual splitting.
<cut/> tag you add. Insert the tag from the toolbar with a single click, or type it manually. Add a name attribute for a custom filename — any alphabet works, including accented Latin, Cyrillic, Chinese, and Arabic.
Start with an empty editor
The cursor blinks at the start of the line — ready for your script.
Type your text and place the cursor at a split point
Write your script — paragraphs, chapters, podcast intros, whatever your project needs. Then place the cursor where one segment should end and the next begins.
Click the Cut button in the toolbar
The Cut button sits in the editor toolbar — the small scissors icon next to SSML. One click inserts a <cut/> tag at the cursor. Repeat at the next split point to add a second cut. Two cuts means three segments.
Recommended limit: 1,000 short segments per project, or 500 for longer ones.
Press Convert — three auto-named segments come out
SpeechGen renders the whole text as a single synthesis (so timing and intonation stay consistent across segments), then slices the audio at every <cut/> tag. Each segment is auto-named from the first words of its text — for example {projectId}_1_chapter-one-the-journey-begins.
Two new buttons appear on the result: Download segments (zip) grabs all files at once; the chevron toggles a list with per-row download for individual segments.
Want custom filenames? Add a name attribute
Type the name attribute directly inside any <cut/> tag. Use any letters, punctuation, or non-Latin alphabets — accented Latin, Cyrillic, Chinese, Arabic — every Unicode character is preserved in the filename:
The name belongs to the segment that ends at this tag. So name="chapter-1" on the first cut renames the first segment.
Convert again — segments now have your custom names
Press Convert. The same text is re-rendered (or pulled from cache if unchanged), then sliced again — this time the named cuts produce chapter-1.mp3 and chapter-2.mp3 instead of auto-generated names.
If you change a filename and re-voice the same text, SpeechGen recognises the unchanged audio and doesn't spend extra limits.
Download all segments at once or one by one
The result panel shows two download options: ① Download segments (zip) grabs every file at once — perfect for batch workflows. ② The chevron toggles a list of every segment with its own per-row download button.
Each segment with the name you set
The list shows every segment with its duration and a download button. Custom-named segments — like the highlighted chapter-1 and chapter-2 rows — keep the exact name you typed. Auto-named segments take their name from the first words of the text.
Default filename pattern: {projectId}_{position}_{transliterated-text}. With name="...", the filename is exactly what you typed.
The result
Listen to the synthesis that came out — three audiobook chapters in one render, sliced into separate files:
Three chapters · Amber
Custom filenames in detail
The name attribute on a cut tag becomes that segment's output filename. The tag must sit at the end of the segment whose name you want to set.
You can use spaces, hyphens, underscores, and any Unicode letters. If you don't add a name, SpeechGen auto-generates one from the first words of the segment.
If you re-voice the same text after changing only the filenames, the system reuses the existing audio and doesn't deduct extra limits — handy when you're iterating on naming for a large batch.
Segment limits
Up to 1,000 short segments or 500 longer segments per voiceover project. If you see empty segments in the output, you've hit the limit — split your text into two separate projects to recover.
When to use it
Anywhere you'd otherwise hand-split audio in Audacity or your DAW. A non-exhaustive list:
Education
- Audiobook chapters or paragraph blocks
- Audio lessons split by topic or sub-topic
- Foreign-language modules (vocabulary, dialogue, drills)
- Audio quizzes split by question
- Lectures, seminars, distance-learning blocks
Media & entertainment
- Podcast intros, segments, sponsor reads, outros — each as a clean file
- Movie quotes batch-voiced from a Google Sheet
- Per-scene audio for radio plays, theatrical productions, indie games
- Trailer voice-overs in multiple lengths from one render
Tourism & travel
- Audio guides for tourist routes (one file per landmark)
- In-flight safety announcements per segment
- Hotel welcome messages by service
- National-park trail commentary
Business & health
- Employee onboarding modules
- Product or service walkthrough files
- Customer-review or testimonial reels (one per quote)
- Guided breathing or meditation sessions split by phase
- Patient-rehabilitation exercise audio per movement
Pair this with a Google Sheets template if you need to bulk-name dozens of segments at once.