stepfun-asr

An agent skill by daymade, from daymade/claude-code-skills. Tags: api, audio, ml, stepfun, transcription.

What it does

Transcribe audio with StepFun's stepaudio-3-asr-max — an SSE endpoint (NOT /v1/audio/transcriptions), single call handles long audio with no client-side chunking. Use when transcribing Chinese / English audio with StepFun, when long-form recordings (5-30 min) need to land in one request, when migrating from step-asr / step-asr-1.1 / stepaudio-2.5-asr, or when hitting the misleading `model stepaudio-3-asr-max not supported` error (which actually means wrong endpoint). Triggers on 阶跃 ASR, StepFun ASR, stepaudio-3-asr-max, stepaudio-2.5-asr, 转录, 语音识别, 长音频转写, 语音转文字. For TTS with the sibling stepaudio-3-tts model, use the stepfun-tts skill instead.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add daymade/claude-code-skills --skill stepfun-asr

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/daymade/claude-code-skills
cp -r claude-code-skills/daymade-audio/stepfun-asr ~/.claude/skills/stepfun-asr

Safety box score

Not rated yet. A safety box score grades what a skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. Anyone can request one from this page; it is saved for everyone. How the score works.

Source

Repository

daymade/claude-code-skills (all skills from this repository)

Path

daymade-audio/stepfun-asr/SKILL.md

Branch

main

Updated

2026-09-20

Related skills

• stepfun-tts — 5-tts` (blind-judged better on neutral/emotive preset voices), `stepaudio-3-tts` for whisper & inline-prosody cases (where it won the same blind test).
• asr-transcribe-to-text — Transcribe audio/video to speaker-labeled text — who-said-what by default, plain-text opt-out; MLX-local on Apple Silicon or remote; local files, media URLs.
• amazon-bedrock — Builds generative AI applications on Amazon Bedrock.
• llm-eval-harness — Test/evaluate any LLM behind an OpenAI- or Anthropic-compatible endpoint: availability (max_tokens-aware).
• fine-tuning-expert — Use when fine-tuning LLMs, training custom models, or adapting foundation models for specific tasks.
• agents-connect — Use when connecting your agent to external APIs, tools, or services via Gateway, or restricting tool access with Cedar policies.