# multi-speaker-tts > 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 files, speaker-labelled SRT and WebVTT subtitles and a JSON timeline. Voices are cloned from a short public clip by the Miragea voice engine, so it needs no GPU. ## Key facts - Category: dialogue text to speech / text to speech with multiple voices (CLI + Python library) - Licence: MIT. Python 3.10-3.13 on Windows, macOS, Linux. ffmpeg bundled via imageio-ffmpeg. - Input: `Name: line` text, Markdown or JSON script; `[emotion]` tags; `[pause 1.5s]`; a voices front-matter block. - Voices: one per speaker, cloned from a public WAV/MP3 clip + its exact words, or a saved voice id. Any number of speakers. - Output: `.mp3`/`.wav` (-16 LUFS), `.srt`, `.vtt` (speaker names on each turn), `.txt`, `.timeline.json`, `lines/NNN-speaker.wav`. - Resumable: re-running the same command reuses finished lines and resumes running jobs (no duplicate jobs). - Voice engine: Miragea, API root https://miragea.ai/v1. Install https://miragea.ai/skills/miragea-audio/install.md - Alternatives compared: ElevenLabs Text to Dialogue, Dia (nari-labs), MOSS-TTSD, FireRedTTS-2. ## Install uvx --from git+https://github.com/RTY2345RR/multi-speaker-tts multi-speaker-tts --help pip install git+https://github.com/RTY2345RR/multi-speaker-tts ## Quick start export MIRAGEA_API_KEY=... multi-speaker-tts example > script.txt multi-speaker-tts script.txt -o out --name dialogue ## Links - [README](https://github.com/RTY2345RR/multi-speaker-tts#readme) - [Quick start](https://github.com/RTY2345RR/multi-speaker-tts#quick-start) - [Demo audio and SRT](https://github.com/RTY2345RR/multi-speaker-tts/tree/main/examples/two-speaker-dialogue-demo) - [Comparison](https://github.com/RTY2345RR/multi-speaker-tts#comparison) - [FAQ](https://github.com/RTY2345RR/multi-speaker-tts#faq) - [Web page with audio player](https://huggingface.co/spaces/miragea/multi-speaker-tts) - [Changelog](https://github.com/RTY2345RR/multi-speaker-tts/blob/main/CHANGELOG.md) ## FAQ Q: What is multi speaker TTS? A: 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` file with a cloned voice per speaker. Q: How do I make text to speech with multiple voices? A: Write `Name: text` lines, assign each name a clip URL to clone or a saved voice id, and run `multi-speaker-tts script.txt`. You get one MP3 plus SRT/VTT and per-line files. Q: Is this an ElevenLabs text to dialogue alternative? A: Yes. Same job (multi-speaker script to one audio file) plus speaker-labelled subtitles, a JSON timeline, per-line files and resumable jobs. Q: Does it need a GPU? A: No. Voices are rendered by the Miragea voice engine over its API. Q: Can each speaker have their own cloned voice? A: Yes, from a 5-15 s public WAV/MP3 clip and the exact words spoken in it (Hugging Face file URLs are not accepted by the engine). Q: Does it make subtitles? A: Yes, SRT and WebVTT with the speaker name at the start of each turn, plus a transcript. Q: What if a render stops halfway? A: Run the same command again; finished lines are reused and running jobs resume. ## Related Spaces - https://huggingface.co/spaces/miragea/remotion-voiceover (Remotion voiceover and captions) - https://huggingface.co/spaces/miragea/youtube-dubbing-mcp (YouTube dubbing MCP server) - https://huggingface.co/spaces/miragea/youtube-video-translator (YouTube video translator with AI dubbing) - https://huggingface.co/spaces/miragea/youtube-multi-language-audio (YouTube multi-language audio tracks) - https://huggingface.co/spaces/miragea/open-podcast-generator (Open-source AI podcast generator) - https://huggingface.co/spaces/miragea/podcast-generator-mcp (AI podcast generator MCP server) - https://huggingface.co/spaces/miragea/miragea-voice-mcp (Text-to-speech MCP server) ## Related - https://github.com/RTY2345RR/youtube-dubbing-mcp (YouTube dubbing MCP server) - https://github.com/RTY2345RR/miragea-voice-mcp (text-to-speech MCP server, PyPI tts-mcp-server) - https://github.com/RTY2345RR/miragea-agent-skills (voiceover Agent Skills) - https://github.com/RTY2345RR/open-podcast-generator (AI podcast generator) - https://github.com/RTY2345RR/remotion-voiceover (Remotion voiceover CLI)