Skip to main content
skill CX Codex

speech-to-text · 101-skills — Claude Code skill for audio transcription

speech-to-text is a Claude Code skill that runs Whisper and ElevenLabs through the belt CLI to transcribe audio and prepare video subtitles.

no license 242 installs ★ 31 updated 2026-08-31

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add 101-skills/superpowers --skill speech-to-text -a claude-code
Open on GitHub

What speech-to-text does

The skill sends audio to transcription models through the belt CLI. Results come back as JSON containing the transcript and detected language; timestamped segments are available when requested. Whisper V3 Large can also translate audio when the task is set to translate.

Available models include ElevenLabs Scribe v2, Fast Whisper Large V3, and Whisper V3 Large. You need the belt CLI and an inference.sh login via belt login; without the CLI, the skill’s commands cannot run. For subtitles, request timestamps and pass the transcript to infsh/caption-videos.

What it helps with

  • transcribe a meeting recording
  • get timestamped text from a podcast
  • translate audio into English
  • prepare a transcript for video captions

Example prompts for the agent

  • Transcribe the audio at https://meeting.mp3 into text.
  • Create a transcript with timestamps for https://podcast.mp3.
  • Translate the audio at https://french-audio.mp3 into English.

How to install speech-to-text

1Skills CLI (Vercel)

npx skills add 101-skills/superpowers --skill speech-to-text -a claude-code

2Manually: put the files in place

~/.claude/skills/speech-to-text/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/speech-to-text/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about speech-to-text

How do I install speech-to-text in Claude Code?

Run npx skills add 101-skills/superpowers --skill speech-to-text -a claude-code. The skill file is placed at ~/.claude/skills/speech-to-text/SKILL.md.

What does speech-to-text do?

It runs Whisper or ElevenLabs models to turn audio into a transcript. The JSON result includes the text and detected language, with timestamped segments available on request.

What does the skill need to run?

Install the belt CLI and sign in to inference.sh with belt login. Audio translation is available through the translate task.

Similar skills

All skills for AI and ML →

speech-to-text by other authors comes first.

speech-to-text elevenlabs speech-to-text — a Claude Code skill that transcribes audio and video to text using ElevenLabs Scribe v3. Runs when converting media, generating subtitles, or transcribing meetings. CX Codex +1 16,200installs speech-to-text NoizAI speech-to-text converts audio and video files to text with automatic language detection, timestamps, and speaker identification. Requires a Noiz API key. CX Codex +1 198installs speech-to-text martinholovsky speech-to-text guides coding agents to build local speech recognition engines using Faster Whisper with a privacy-first TDD workflow. CX Codex +1 284installs text-to-speech elevenlabs text-to-speech is a Claude Code skill for generating speech with ElevenLabs. It activates for text-to-audio tasks, voiceovers, and voice app development. CX Codex +1 22,900installs azure-speech-to-text calesthio azure-speech-to-text transcribes local audio files into text with word-level timestamps and speaker diarization using Azure AI Speech. Speeds up subtitle creation and audio processing in OpenMontage pipelines. CX Codex +1 913installs azure-text-to-speech calesthio azure-text-to-speech synthesizes neural voiceover audio using the Azure AI Speech REST API. Ideal for OpenMontage pipelines that need multilingual narration voices. CX Codex +1 738installs syncfusion-react-speech-to-text Syncfusion Inc The syncfusion-react-speech-to-text skill guides React implementations of Syncfusion SpeechToText, including microphone input, transcript events, listening control, and accessible voice forms. CX Codex +1 735installs Text to Speech — 20+ Languages, MP3 Output axel-belfort Text to Speech — 20+ Languages, MP3 Output is a Claude Code MCP server that converts text into speech in 20+ languages and returns base64 MP3 audio for voice alerts and spoken UI. CX Codex +1 2,769installs azure-speech-to-text-rest-py Microsoft azure-speech-to-text-rest-py — a skill for short audio transcription via the Azure Speech REST API in Python. It handles speech recognition under 60 seconds without installing the Speech SDK. CX Codex +1 2installs

From the same set81

The whole set →
elevenlabs-sound-effects elevenlabs-sound-effects generates audio effects from text prompts using ElevenLabs and the inference.sh CLI tool. CX Codex +1 244installs ai-product-photography ai-product-photography turns a product brief into commercial images through the inference.sh CLI: studio shots, lifestyle scenes and marketplace cards. CX Codex +1 243installs background-removal background-removal cuts the background out of photos with the BiRefNet model through the belt CLI and returns a transparent PNG. Handy for product shots and headshots. CX Codex +1 243installs tools-ui tools-ui is a skill for building React/Next.js interfaces for agent tool calls, showing progress, approval prompts, and results. It makes tool activity visible before and after execution. CX Codex +1 243installs case-study-writing case-study-writing guides drafting B2B case studies using the STAR method, metric comparisons, customer quotes, and chart visuals. CX Codex +1 242installs affirmations A skill for an agent stuck in a loop: it reads a short set of affirmations, re-checks what it actually knows, and takes one small step. CX Codex +1 242installs chat-ui chat-ui gives your agent ready-made chat building blocks for React and Next.js: container, messages, input field, typing indicator. Useful when the app needs a conversation screen. CX Codex +1 242installs talking-head-production talking-head-production turns a still portrait plus a script into a talking-head clip: the skill walks the agent through portrait checks, voice choice and lipsync calls on the belt CLI. CX Codex +1 241installs

Project sponsors

Companies that keep this analytics open