{"id":44858348,"url":"https://github.com/agentem-ai/izwi","last_synced_at":"2026-03-10T09:01:13.385Z","repository":{"id":335532256,"uuid":"1140713799","full_name":"agentem-ai/izwi","owner":"agentem-ai","description":"On-device Voice AI engine for transcription, TTS, and voice workflows.","archived":false,"fork":false,"pushed_at":"2026-03-05T08:58:45.000Z","size":30280,"stargazers_count":158,"open_issues_count":6,"forks_count":14,"subscribers_count":5,"default_branch":"main","last_synced_at":"2026-03-05T13:09:16.051Z","etag":null,"topics":["asr","audio-inference","local-first","openai-compatible-api","self-hosted-ai","speaker-diarization","speech-to-text","text-to-speech","tts","voice-cloning"],"latest_commit_sha":null,"homepage":"https://www.izwiai.com","language":"Rust","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/agentem-ai.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-01-23T16:53:17.000Z","updated_at":"2026-03-05T13:01:33.000Z","dependencies_parsed_at":"2026-03-02T15:00:52.658Z","dependency_job_id":null,"html_url":"https://github.com/agentem-ai/izwi","commit_stats":null,"previous_names":["agentem-ai/izwi-audio","agentem-ai/izwi"],"tags_count":11,"template":false,"template_full_name":null,"purl":"pkg:github/agentem-ai/izwi","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agentem-ai%2Fizwi","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agentem-ai%2Fizwi/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agentem-ai%2Fizwi/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agentem-ai%2Fizwi/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/agentem-ai","download_url":"https://codeload.github.com/agentem-ai/izwi/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agentem-ai%2Fizwi/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":30328266,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-03-10T05:25:20.737Z","status":"ssl_error","status_checked_at":"2026-03-10T05:25:17.430Z","response_time":106,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":["asr","audio-inference","local-first","openai-compatible-api","self-hosted-ai","speaker-diarization","speech-to-text","text-to-speech","tts","voice-cloning"],"created_at":"2026-02-17T08:03:23.736Z","updated_at":"2026-03-10T09:01:13.321Z","avatar_url":"https://github.com/agentem-ai.png","language":"Rust","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/app-icon.png\" alt=\"Izwi icon\" width=\"140\" /\u003e\n\u003c/p\u003e\n\n\u003ch1 align=\"center\"\u003eIzwi\u003c/h1\u003e\n\n\u003cp align=\"center\"\u003e\u003cstrong\u003eLocal-first audio inference engine for TTS, ASR, and voice AI workflows.\u003c/strong\u003e\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://izwiai.com\"\u003eWebsite\u003c/a\u003e •\n  \u003ca href=\"https://izwiai.com/docs\"\u003eDocumentation\u003c/a\u003e •\n  \u003ca href=\"https://github.com/agentem-ai/izwi/releases\"\u003eReleases\u003c/a\u003e •\n  \u003ca href=\"https://izwiai.com/docs/getting-started\"\u003eGetting Started\u003c/a\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/screenshot.png\" alt=\"Izwi Screenshot\" width=\"800\" /\u003e\n\u003c/p\u003e\n\n---\n\n## Overview\n\nIzwi is a privacy-focused audio AI platform that runs entirely on your machine. No cloud services, no API keys, no data leaving your device.\n\n**Core capabilities:**\n\n- **Voice Mode** — Real-time voice conversations with AI\n- **Text-to-Speech** — Generate natural speech from text\n- **Speech Recognition** — Convert audio to text with high accuracy\n- **Speaker Diarization** — Identify and separate multiple speakers\n- **Voice Cloning** — Clone any voice from a short audio sample\n- **Voice Design** — Create custom voices from text descriptions\n- **Forced Alignment** — Word-level audio-text alignment\n- **Chat** — Text-based AI conversations\n\nThe server exposes OpenAI-compatible API routes under `/v1`.\n\n---\n\n## Quick Install\n\n### macOS\n\nDownload the latest `.dmg` from [GitHub Releases](https://github.com/agentem-ai/izwi/releases):\n\n1. Open the `.dmg` file\n2. Drag **Izwi.app** to Applications\n3. Launch Izwi\n\n### Linux\n\n```bash\nwget https://github.com/agentem-ai/izwi/releases/latest/download/izwi_amd64.deb\nsudo dpkg -i izwi_amd64.deb\n```\n\n### Windows\n\nDownload and run the installer from [GitHub Releases](https://github.com/agentem-ai/izwi/releases).\n\n\u003e **Full installation guides:** [macOS](https://izwiai.com/docs/installation/macos) • [Linux](https://izwiai.com/docs/installation/linux) • [Windows](https://izwiai.com/docs/installation/windows) • [From Source](https://izwiai.com/docs/installation/from-source)\n\n---\n\n## Quick Start\n\n### 1. Start the server\n\n```bash\nizwi serve\n```\n\nOpen `http://localhost:8080` in your browser.\n\n### 2. Download a model\n\n```bash\nizwi pull Qwen3-TTS-12Hz-0.6B-Base\n```\n\n### 3. Generate speech\n\n```bash\nizwi tts \"Hello from Izwi!\" --output hello.wav\n```\n\n### 4. Transcribe audio\n\n```bash\nizwi pull Qwen3-ASR-0.6B\nizwi transcribe audio.wav\n```\n\nLong-form ASR is handled automatically: Izwi now chunks long recordings,\nstitches overlapping transcripts, and returns a full transcript instead of\nonly the first model window.\n\nOptional tuning knobs:\n\n```bash\nIZWI_ASR_CHUNK_TARGET_SECS=24\nIZWI_ASR_CHUNK_MAX_SECS=30\nIZWI_ASR_CHUNK_OVERLAP_SECS=3\n```\n\n---\n\n## Supported Models\n\n| Category | Models |\n|----------|--------|\n| **TTS** | Qwen3-TTS (0.6B, 1.7B), LFM2-Audio |\n| **ASR** | Qwen3-ASR (0.6B, 1.7B), Parakeet TDT |\n| **Diarization** | Sortformer 4-speaker |\n| **Chat** | Qwen3 (0.6B, 1.7B), Gemma 3 (1B, 4B) |\n| **Alignment** | Qwen3-ForcedAligner |\n\nRun `izwi list` to see all available models.\n\n\u003e **Full model documentation:** [Models Guide](https://izwiai.com/docs/models)\n\n---\n\n## Documentation\n\n| Resource | Link |\n|----------|------|\n| **Getting Started** | [izwiai.com/docs/getting-started](https://izwiai.com/docs/getting-started) |\n| **Installation** | [izwiai.com/docs/installation](https://izwiai.com/docs/installation) |\n| **Features** | [izwiai.com/docs/features](https://izwiai.com/docs/features) |\n| **CLI Reference** | [izwiai.com/docs/cli](https://izwiai.com/docs/cli) |\n| **Models** | [izwiai.com/docs/models](https://izwiai.com/docs/models) |\n| **Troubleshooting** | [izwiai.com/docs/troubleshooting](https://izwiai.com/docs/troubleshooting) |\n\n---\n\n## License\n\nApache 2.0\n\n## Acknowledgments\n\n- [Qwen3-TTS](https://github.com/QwenLM/Qwen3-TTS) by Alibaba\n- [Parakeet](https://huggingface.co/nvidia/parakeet-tdt-0.6b-v2) by NVIDIA\n- [Gemma](https://ai.google.dev/gemma) by Google\n- [LFM2-Audio](https://www.liquid.ai/) by Liquid AI\n- [HuggingFace Hub](https://huggingface.co/) for model hosting\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fagentem-ai%2Fizwi","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fagentem-ai%2Fizwi","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fagentem-ai%2Fizwi/lists"}