Convert audio files to text accurately using Qwen3 speech recognition models.
Install
mkdir -p .claude/skills/skills-second-state && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/10964" && unzip -o skill.zip -d .claude/skills/skills-second-state && rm skill.zipInstalls to .claude/skills/skills-second-state
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Transcribe speech from audio files to text.Key capabilities
- →Transcribes audio files to text
- →Supports multiple audio formats
- →Uses Qwen3-ASR model
- →Prints transcription to stdout
How it works
It utilizes the Qwen3-ASR model to process audio files and output the recognized speech as text.
Inputs & outputs
When to use skills
- →Transcribing recorded interviews
- →Converting audio meeting notes to text
- →Processing media files for documentation
About this skill
Qwen3 ASR — Voice Transcription
Transcribe speech from audio files to text.
Binary
{baseDir}/scripts/asr— Speech-to-text transcription.
Models
{baseDir}/scripts/models/Qwen3-ASR-0.6B— Speech recognition model (0.6B parameters).
Transcription
Transcribe an audio file to text.
{baseDir}/scripts/asr \
{baseDir}/scripts/models/Qwen3-ASR-0.6B \
<audio_file>
Parameters
| Parameter | Required | Description |
|---|---|---|
| model_path | Yes | Path to the model directory (0.6B or 1.7B) |
| audio_file | Yes | Path to the audio file (any FFmpeg-supported format) |
Output
Prints the transcribed text to standard output.
Example
{baseDir}/scripts/asr \
{baseDir}/scripts/models/Qwen3-ASR-0.6B \
recording.wav
Supported Audio Formats
Any format supported by FFmpeg: WAV, MP3, M4A, FLAC, OGG, and more. Audio is automatically resampled to 16 kHz mono internally.
Workflow
1. Identify the Audio File
Get the path to the audio file the user wants to transcribe.
2. Run the Command
Run the asr binary with the full paths to the binary and model directory.
{baseDir}/scripts/asr \
{baseDir}/scripts/models/Qwen3-ASR-0.6B \
/path/to/audio.wav
3. Return the Transcription
The transcribed text is printed to stdout. Return it to the user.
When not to use it
- →When the audio file is not supported by FFmpeg
- →When the user needs real-time transcription
Limitations
- →Requires model files
- →Offline processing only
How it compares
It provides a dedicated CLI tool for offline, model-based transcription.
Compared to similar skills
skills side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| skills (this skill) | 0 | 5mo | Review | Beginner |
| hugging-face-evaluation | 0 | 2mo | Review | Intermediate |
| Create SDRF | 0 | 6mo | Review | Beginner |
| report-research | 1 | 6mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by second-state
View all by second-state →You might also like
hugging-face-evaluation
FISCFED9
Add and manage evaluation results in Hugging Face model cards. Supports extracting eval tables from README content, importing scores from Artificial Analysis API, and running custom model evaluations with vLLM/lighteval. Works with the model-index metadata format.
Create SDRF
bigbio
Create a sample-to-data-relationship format (SDRF) file (usually from another type of samplesheet)
report-research
numerai
Write a complete Numerai experiment report in experiment.md (abstract, methods, results tables, decisions, next steps) and generate/link the standard show_experiment plot(s). Use after running any Numerai research experiments, or when a user asks for a “full report”, “write up”, “experiment.md update”, or “generate the standard plot”.
document-pro
bighardperson
文档处理技能 - 让 AI 能够读取、解析、提取 PDF、DOCX、PPT 等文档的关键信息。当用户要求分析文档、提取内容、总结报告时触发此技能。
modeling-project-structure
shengzhou-nwpu
>-
quant-analyst
zenobi-us
Expert quantitative analyst specializing in financial modeling, algorithmic trading, and risk analytics. Masters statistical methods, derivatives pricing, and high-frequency trading with focus on mathematical rigor, performance optimization, and profitable strategy development.