Text to speech MP3: paste a script or drop SRT and TXT files, get an AI voice
Only need a voice for a podcast, an audiobook or a video you edit in CapCut or Premiere? The Text to speech window in DubLocal Studio reads the text you paste, or a whole batch of SRT and TXT files, into MP3: one MP3 per file with the same name, plus an SRT that matches the new voice if you want one.
Card payment for international customers is coming soon.
Updated October 2026 · The DubLocal team
How to turn a script into an MP3
- 01
Open Text to speech
Click Text to speech in Studio's top bar; no video or project needed.
- 02
Paste text or drop files
Paste the script in the Paste text tab, or drop several .srt and .txt files in the SRT/TXT files tab.
- 03
Choose a voice
Pick the language, voice, speed and the pause between sentences; listen to the voice before reading.
- 04
Make the MP3
Click Make voice → MP3. Studio reads each sentence, joins them into an MP3 and saves it to your folder.
Text, SRT or TXT, all read aloud
Paste text: each sentence or line is read in turn, and a blank line starts a new paragraph with a longer pause. Name the MP3, or leave the name empty and Studio uses the first few words.
SRT and TXT files: drop many at once and each one becomes an MP3 with the same name. The voice and speed apply to every file in that run; a line that fails is skipped and Studio tells you how many it skipped.
Reading pace and the matching SRT
Sentences are read one after another with the pause you choose, not on the timings of the original SRT. Turn on the matching .srt option to get subtitles timed to the new voice, ready for CapCut or Premiere.
Need a voice timed to an existing video? Use SRT to voice in a dubbing project instead: every line is read inside its own time slot.
10 reading languages
Vietnamese, English, German, French, Spanish, Indonesian, Japanese and Korean read on the CPU, no graphics card needed. Thai and Chinese read with OmniVoice: install the multilingual pack under Install AI models; an NVIDIA card makes it fast, on the CPU it is very slow.
- Podcasts, audiobooks and lessons read from text
- Narration for videos you edit in CapCut or Premiere
- A whole folder of SRT and TXT files turned into MP3s in one run
Watch DubLocal Studio from start to finish (Vietnamese audio)
The tutorial covers installing, opening a video, transcribing, translating, casting, generating voices and exporting MP4.
Frequently asked questions
Do I need a video?
No. Text to speech only needs text or SRT and TXT files and exports straight to MP3.
Can it read many files at once?
Yes. In the SRT/TXT files tab, drop up to 200 files at a time; each becomes an MP3 with the same name in the folder you choose.
Does the MP3 follow the SRT timings?
No: the sentences are read one after another. Turn on the matching .srt option for new subtitles that follow the voice; to match an existing video, use SRT to voice in a dubbing project.
Does it work on a slower PC?
Yes for Vietnamese, English, German, French, Spanish, Indonesian, Japanese and Korean: those voices run on the CPU. Thai and Chinese need an NVIDIA card to read quickly.
Which plan do I need?
DubLocal Studio on a Windows PC and an active license.