Claude Skill MartBrowse skillsWhat is a Skill?
 
← Back to list

StepFun ASR (stepaudio-2.5-asr) Transcription

Transcribe up to 30 minutes of audio in a single API call using StepFun's stepaudio-2.5-asr SSE endpoint.

Image & VideoIntermediate★ 1,323⑂ 212AI score 8/10Last updated: Aug 8, 2026

What it does

  • Transcribes audio files with StepFun's stepaudio-2.5-asr model (Chinese and English focused).
  • A 32K context window means 5–30 minute recordings land in one request with no client-side chunking — 17.4 minutes of audio transcribes in about 10 seconds (85–101× real time).
  • The bundled scripts/asr_transcribe.py handles base64 encoding, the nested JSON body, and SSE stream parsing; add --json for token usage and timing data.
  • Documents the three traps that eat hours: the wrong endpoint returning a misleading model ... not supported error, "Plan" keys silently failing on audio endpoints, and mid-stream error SSE events.
  • Supports mp3 / wav / ogg / opus / pcm; transcode mp4, m4a or webm with ffmpeg first.

Who it's for

  • Anyone regularly transcribing Chinese or English meetings, interviews, or podcasts.
  • Developers migrating from step-asr / step-asr-1.1 who want to know if the speedup is worth it.
  • Engineers stuck on the model stepaudio-2.5-asr not supported error.
  • Teams wanting to delete their audio-chunking pipeline.

Examples

  1. 20-minute standup recording: python3 scripts/asr_transcribe.py meeting.mp3 gives you the full text, then hand it to a meeting-minutes skill.
  2. English interview: run with --language en and --json to capture usage.total_tokens for cost tracking.
  3. One-hour lecture: exceeds the ~30 minute ceiling, so split it with ffmpeg, transcribe each half, and stitch the output into a subtitle draft.

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the daymade-audio/stepfun-asr folder from the GitHub repo daymade/claude-code-skills into my ~/.claude/skills/stepfun-asr/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/daymade/claude-code-skills.git /tmp/ccs && mkdir -p ~/.claude/skills && cp -r /tmp/ccs/daymade-audio/stepfun-asr ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and clone the repository: 
    git clone https://github.com/daymade/claude-code-skills.git /tmp/ccs
  2. Copy the skill into your Claude Code skills folder: 
    mkdir -p ~/.claude/skills && cp -r /tmp/ccs/daymade-audio/stepfun-asr ~/.claude/skills/
  3. Sign up at https://platform.stepfun.com/ and create a key under API Keys. Make sure it is a Normal key — "Plan" subscription keys cannot call audio endpoints and fail with an unhelpful 4xx.
  4. Export the key: 
    export STEPFUN_API_KEY="your_key_here"
    Add the line to ~/.zshrc or ~/.bashrc to persist it.
  5. Confirm Python 3 and ffmpeg are installed (ffmpeg is needed to convert mp4/m4a into mp3 or ogg).
  6. Restart Claude Code and ask something like "transcribe this recording with StepFun ASR" — the skill triggers automatically.
View source on GitHub ↗License: MIT