talkiesSelf-hosted OpenAI-compatible speech service. /v1/audio/transcriptions fronts 12 open ASR models (Whisper, Parakeet, Nemotron-3.5-ASR, Canary, Sherpa-ONNX, Vosk); /v1/audio/transcriptions/stream accepts live PCM over WebSocket. /v1/audio/speech fronts 2 TTS engines / 3 backends — Kokoro-82M (41 baked voices, PyTorch + ONNX runtimes) and the CUDA-only Qwen3-TTS family (voice cloning, preset speakers, voice design). Stereo diarization, URL fetching, MCP endpoint, bearer auth.
Install via ClawdBot CLI:
clawdbot install psyb0t/talkiesGrade Limited — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Sends data to undocumented external endpoint (potential exfiltration)
POST → http://localhost:8000/unloadCalls external URL not in known-safe list
https://github.com/psyb0t/docker-talkiesUses known external API (expected, informational)
api.openai.comAudited May 29, 2026 · audit v1.0
Usage Guide
Loading usage data… refresh in a few seconds.
Scored May 29, 2026
Local speech-to-text with the Whisper CLI (no API key).
Transcribe audio files with speaker diarization (who speaks when). Supports 100+ languages, automatic language detection, and timestamps. Use for meetings, interviews, podcasts, or voice messages. Requires AssemblyAI API key.
Secure, offline, OpenAI-compatible local Whisper ASR endpoint for OpenClaw. Features faster-whisper (large-v3-turbo), built-in privacy with no cloud telemetr...
Turn your AI into JARVIS. Voice, wit, and personality — the complete package. Humor cranked to maximum.
Text-to-speech conversion using node-edge-tts npm package for generating audio from text. Supports multiple voices, languages, speed adjustment, pitch contro...
使用 poocr 库识别发票并导出 Excel。当用户需要识别增值税发票、批量处理发票文件或提取发票信息到 Excel 时调用此技能。