speech-to-text

Name: speech-to-text
Availability: InStock
Author: ElevenLabs

Transcribe audio to text using ElevenLabs Scribe v2.

412|55|Updated Jan 24, 2026

Search Tags:#real-time #transcription #timestamps #speech-to-text #diarization #Scribe v2

Ready to run

npx skills add https://github.com/elevenlabs/skills --skill speech-to-text

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: speech-to-text
Source: https://github.com/elevenlabs/skills/tree/main/speech-to-text
Command: npx skills add https://github.com/elevenlabs/skills --skill speech-to-text

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill converts spoken content from audio or video into searchable, editable transcripts, enabling subtitles, meeting minutes, and data extraction.

Core Features & Use Cases

High-accuracy transcription: supports 90+ languages, word-level timestamps, and speaker diarization to distinguish multiple voices.
Versatile workflows: supports batch transcription with language hints and real-time streaming for live captions and subtitles.
Practical scenarios: transcribe meetings, interviews, podcasts, or lectures and generate searchable transcripts with speaker labels.

Quick Start

Use the speech-to-text skill to transcribe an audio file with model_id "scribe_v2" and optionally enable timestamps and diarization for richer output.

Frequently Asked Questions about speech-to-text

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I transcribe audio to text with word-level timestamps and speaker diarization?▼

You can transcribe audio to text with word-level timestamps and speaker diarization by submitting audio files to the Scribe v2 model. This Skill provides usage guidance for enabling these features across Python, JavaScript/TypeScript, and cURL workflows.

Can I use real-time streaming transcription for live captions and subtitles?▼

Yes, real-time streaming transcription is supported for live captions and subtitles. Scribe v2 handles both batch processing and real-time audio streams to generate timestamped text for immediate display.

Does speech-to-text transcription work with multiple languages and speaker identification?▼

Speech-to-text transcription supports over 90 languages and includes speaker diarization to distinguish multiple voices. You can provide language hints during batch processing to improve accuracy for specific audio content.

What do I need to set up before transcribing audio files with Scribe v2?▼

You need an ElevenLabs API key (ELEVENLABS_API_KEY) and internet access to transcribe audio files with Scribe v2. Installation and usage instructions are provided in the References section for Python, JavaScript/TypeScript, and cURL environments.

What is the best way to generate meeting minutes from recorded conversations?▼

To generate meeting minutes from recorded conversations, use batch transcription with Scribe v2 and enable speaker diarization. This produces searchable transcripts with speaker labels and word-level timestamps for identifying key discussion points.

Are there limitations when processing audio for transcription across different programming languages?▼

Audio transcription with Scribe v2 requires internet access and a valid API key, with implementation examples limited to Python, JavaScript/TypeScript, and cURL. The Skill does not process audio offline or without proper authentication credentials.