Menu
Transcribe speech audio to text using the local LFM2.5 model on the GPU.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Use open weight models (LLM, image, audio) with open source agents on Mac, Linux, and Windows.
Add a new model to the a2go registry โ both GGUF (Linux/Windows) and MLX (macOS) variants, with full testing on Runpod.
Generate images from text prompts using the local FLUX.2 model on the GPU.
Convert text to speech audio using the local LFM2.5 model on the GPU.
Based on SOC occupation classification
| name | a2go-speech-to-text |
| description | Transcribe speech audio to text using the local LFM2.5 model on the GPU. |
| metadata | {"openclaw":{"emoji":"๐ค","requires":{"bins":["a2go"]}}} |
Transcribe speech audio to text using the local LFM2.5-Audio model running on the GPU.
Use the exec tool to run:
a2go tool speech-to-text /path/to/audio.wav
Options:
--output: save transcript to file (default: prints to stdout)Examples:
a2go tool speech-to-text /tmp/recording.wava2go tool speech-to-text /tmp/voice.wav --output /tmp/transcript.txt