Multi speaker TTS: dialogue text to speech with a cloned voice per speaker

multi-speaker-tts is an open-source Python library and command-line tool for multi speaker TTS: it turns a plain-text dialogue script (Name: line) into one stitched MP3 or WAV with a different cloned voice for every speaker. It also writes per-line audio, speaker-labelled SRT and WebVTT subtitles and a JSON timeline, with no GPU. Code on GitHub (MIT).

multi-speaker-tts demo: a two-speaker dialogue script rendered to one MP3 with a cloned male voice (Leo) and a cloned female voice (Maya), with speaker-labelled SRT subtitles

Listen: two speakers, one male and one female cloned voice

two-speaker-dialogue-demo.mp3 (10.3 s, unedited output)

SRT · VTT · JSON timeline

Transcript
Leo (cloned male voice, 0.00–5.93 s): This whole conversation came from one plain text script, with a cloned voice for each of us.
Maya (cloned female voice, [excited], 6.43–10.02 s): And it wrote the subtitles too, with our names on every line!

Quick start

uvx --from git+https://github.com/RTY2345RR/multi-speaker-tts multi-speaker-tts --help
export MIRAGEA_API_KEY=...        # https://miragea.ai/workspace/api
multi-speaker-tts example > script.txt
multi-speaker-tts script.txt -o out --name dialogue

Using an AI agent? Install https://miragea.ai/skills/miragea-audio/install.md

  1. Install: Run uvx --from git+https://github.com/RTY2345RR/multi-speaker-tts multi-speaker-tts --help, or pip install the GitHub repo.
  2. Set the voice-engine key: Create a key at https://miragea.ai/workspace/api and export MIRAGEA_API_KEY.
  3. Write the script: One line per turn as Name: text, with a voices block giving each speaker a clip URL to clone and its exact words.
  4. Render: Run multi-speaker-tts script.txt -o out to get one MP3, SRT, VTT, a transcript and a JSON timeline.

How it works

How multi-speaker-tts works: dialogue script, one voice per speaker, voice each line with Miragea voice cloning, stitch with pauses and loudness normalization, export MP3, SRT, VTT and JSON timeline

Dialogue text to speech features

Multi speaker TTS comparison

multi-speaker-ttsElevenLabs Text to DialogueDiaMOSS-TTSDFireRedTTS-2
TypeOpen-source CLI + libraryHosted APIOpen-weights modelOpen-weights modelOpen-weights model
Local GPUNoNoYesYesYes
SpeakersAny numberMultiple21 to 5Up to 4
Speaker-labelled SRT/VTTYesNot documentedNoNoNo
ResumeYesNot documentedNoNoNo
LicenceMITProprietaryApache-2.0Apache-2.0Apache-2.0

Other projects: facts from their public READMEs, November 2026. "Not documented" means we could not find it.

FAQ

What is multi speaker TTS?

Multi speaker TTS is text to speech that renders several distinct voices in one recording, usually from a script where each line names its speaker. multi-speaker-tts does this from a Name: line text file with a cloned voice per speaker.

How do I make text to speech with multiple voices?

Write one line per turn as Name: text, assign each name a clip URL to clone or a saved voice id, and run multi-speaker-tts script.txt. You get one stitched MP3 plus subtitles and per-line files.

Is this an ElevenLabs text to dialogue alternative?

Yes. It does the same job and adds speaker-labelled SRT/VTT, a JSON timeline, per-line files and resumable jobs, with voice cloning per speaker.

Does multi-speaker-tts need a GPU?

No. Voices are rendered by the Miragea voice engine over its API, so it runs on any laptop.

Can I clone a different voice for each speaker?

Yes. Give each speaker a 5-15 s public WAV/MP3 clip and the exact words in it. Hugging Face file URLs are not accepted by the engine.

Can I add emotions, laughs or pauses?

Yes. Write [laughs], [whispering], (excited) or Maya [sad]: ..., and [pause 2s] on its own line or inside a line.

Does it make subtitles for the dialogue?

Yes. Every render writes SRT and WebVTT with the speaker name at the start of each turn, plus a plain transcript.

What happens if a render stops halfway?

Run the same command again. Finished lines are reused and running jobs resume, so no line is voiced twice.

Related open-source voice tools

Voices by the Miragea voice engine (https://miragea.ai/v1). Demo reference voices: CMU ARCTIC (bdl, slt), © Carnegie Mellon University.