Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

StepFun stepaudio-2.5-tts Speech Synthesis

Generate Chinese/Japanese speech with StepFun's contextual TTS, controlling emotion via natural-language instructions and inline () prosody cues.

Image & VideoIntermediate★ 1,393⑂ 219AI score 8/10Last updated: Sep 15, 2026

What it does

  • Generates Chinese and Japanese speech using StepFun's stepaudio-2.5-tts model.
  • Replaces fixed emotion labels (voice_label) with a natural-language instruction (≤200 chars) for global tone, plus inline () directives for pauses, breath, stress, and mid-sentence mood pivots.
  • Ships a full migration playbook from step-tts-2: emotion→instruction mapping table, non-destructive A/B output directories, and speed/quality trade-off data.
  • Documents real failures and fixes: voice_label is not supported for v2 models, censorship_block on sensitive words, silent truncation past the 1000-char hard cap.
  • Bundled scripts cover single-line synthesis, JSONL batch runs, and A/B comparison between two models.

Who it's for

  • Game/app developers batch-producing Chinese voice lines.
  • Teams migrating an existing step-tts-2 pipeline to the 2.5 model.
  • Narration and character-dub creators who need fine prosody control.

Examples

  1. python3 scripts/tts_generate.py --text "你好" --out /tmp/hello.mp3 --instruction "温暖的希望感" for a warm single line.
  2. Inline direction: "你好(停顿一下)我是蕾格(轻声)今天(加重)的天气真不错。" — pause, soften, then emphasize.
  3. Batch 29 dialogue lines; when 2 hit censorship_block, log the IDs, keep going, and fall back to step-tts-2 only for those lines.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/daymade/claude-code-skills/HEAD/daymade-audio/stepfun-tts/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

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-tts folder from the GitHub repo daymade/claude-code-skills into my ~/.claude/skills/stepfun-tts/.
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-tts ~/.claude/skills/

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

  1. Clone the repo: git clone https://github.com/daymade/claude-code-skills.git
  2. Copy the skill into place: mkdir -p ~/.claude/skills && cp -r claude-code-skills/daymade-audio/stepfun-tts ~/.claude/skills/
  3. Create an API key at https://platform.stepfun.com/ → API Keys. Choose a Normal key — Plan keys fail silently on audio endpoints.
  4. Export it: export STEPFUN_API_KEY="your-key" (add to ~/.zshrc or ~/.bashrc to persist).
  5. Restart Claude Code, then ask something like "synthesize this Chinese line with a whispered, sad tone".
  6. Requires Python 3 and internet access; pricing is roughly ¥5.8 per 10k characters, so estimate cost before large batches.
View source on GitHub ↗License: MIT