{"id":30863173,"url":"https://github.com/mlang/llm-tts","last_synced_at":"2025-09-07T18:07:21.448Z","repository":{"id":300995989,"uuid":"1007833008","full_name":"mlang/llm-tts","owner":"mlang","description":"Streaming playback for the OpenAI Text-to-Speech API","archived":false,"fork":false,"pushed_at":"2025-08-09T17:19:02.000Z","size":13,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-08-09T19:13:22.817Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/mlang.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-06-24T15:44:59.000Z","updated_at":"2025-08-09T17:19:05.000Z","dependencies_parsed_at":"2025-06-24T16:49:04.381Z","dependency_job_id":"27e214fe-ef6e-428f-9645-66501ec075f6","html_url":"https://github.com/mlang/llm-tts","commit_stats":null,"previous_names":["mlang/llm-tts"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/mlang/llm-tts","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mlang%2Fllm-tts","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mlang%2Fllm-tts/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mlang%2Fllm-tts/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mlang%2Fllm-tts/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/mlang","download_url":"https://codeload.github.com/mlang/llm-tts/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mlang%2Fllm-tts/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":274073753,"owners_count":25217774,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-09-07T02:00:09.463Z","response_time":67,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-09-07T18:07:17.444Z","updated_at":"2025-09-07T18:07:21.433Z","avatar_url":"https://github.com/mlang.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# llm-tts\n\n\u003e Streaming playback for Text-to-Speech APIs and local models\n\nProvides a subcommand `llm tts` as well as a `tts` tool.\n\nUses FFmpeg or GStreamer for real-time playback.\n\n## Installation\n\nIf you have FFmpeg installed:\n\n```bash\nllm install git+https://github.com/mlang/llm-tts\n```\n\nIf you prefer to use GStreamer:\n\n```bash\nsudo apt install cairo-1.0 gstreamer-1.0 girepository-2.0\nllm install 'git+https://github.com/mlang/llm-tts#[gstreamer]'\n```\n\n## Usage\n\n```bash\nllm tts 'Hello, World!'\nllm \"A short poem\" | llm tts --instructions poetic\n```\n\nBy default, `llm-tts` will try to use FFmpeg for real-time playback.\nIf FFmpeg doesn’t work for you, try GStreamer instead:\n\n```bash\nllm tts --play gstreamer:alsasink \"Advanced Linux Sound Architecture\"\n```\n\nIf you want the chat model to be able to pass instructions along to the tts model, you can do so with a schema specification and the `--json` flag:\n\n```bash\nllm --schema \"text: Text to speak, instructions: How to speak the given text\" --save tts\nllm -t tts \"A lovely poem\" | llm tts --json\n\n\n## Supported TTS back-ends\n\n### OpenAI\n• Model names: `tts-1`, `tts-1-hd`, `gpt-4o-mini-tts`  \n• Requires `OPENAI_API_KEY`.\n\n### ElevenLabs\n\n* Model name: `eleven_multilingual_v2`  \n* `pip install elevenlabs` and set `ELEVENLABS_API_KEY`.\n\n### Hugging Face / transformers (local)\n\n* Model names: `facebook/mms-tts-eng`, `suno/bark`, `suno/bark-small`  \n* `pip install transformers sentencepiece`.\n\n### Piper / Mimic3 (fully offline)\n\n* Model names start with `piper/`, e.g. `piper/en_US-amy-low`  \n* `pip install piper-tts` – the first run auto-downloads the voice file.\n\n### OuteTTS\n\n* Model names: `OuteTTS-1.0-0.6B`, `Llama-OuteTTS-1.0-1B`  \n* `pip install outetts`.\n\n### Silero (offline)\n\n• Model names such as `silero/v3_en`, `silero/v4_ru`, …  \n• `pip install torch` (models are fetched via `torch.hub`).\n\nRun\n\n```bash\nllm tts --list-models\n```\n\nto see every identifier currently available.  \nIf you omit `--model`, the default is **gpt-4o-mini-tts**.\n\n---\n\n## Playback back-ends\n\n### FFmpeg (default on Linux, macOS, Windows)\n\n```bash\n--play ffmpeg:FORMAT:DEVICE   # e.g. --play ffmpeg:alsa:hw:0\n```\n\n### GStreamer (default on other systems)\n\n```bash\n--play gstreamer[:SINK]       # list sinks with --list-sinks\n```\n\nYou can also skip playback and write the result to a file instead:\n\n```bash\nllm tts --output-file out.mp3 \"Save to disk instead of playing\"\n```\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmlang%2Fllm-tts","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmlang%2Fllm-tts","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmlang%2Fllm-tts/lists"}