speech-to-text

Transcribe audio to text using Whisper models via the inference.sh CLI.

4|1|Updated Feb 10, 2026
One-click install
npx skills add https://github.com/Sheshiyer/brandmint-oracle-aleph --skill speech-to-text-sheshiyer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: speech-to-text
Source: https://github.com/Sheshiyer/brandmint-oracle-aleph/tree/main/skills/external/inference-sh/upstream/ab546d072f1e/tools/audio/speech-to-text
Command: npx skills add https://github.com/Sheshiyer/brandmint-oracle-aleph --skill speech-to-text-sheshiyer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Transcribing audio to text enables accessibility, searchability, and content repurposing across meetings, podcasts, and lectures.

Core Features & Use Cases

  • Transcription: Produce accurate transcripts from audio in multiple languages.
  • Translation: Translate transcripts when needed for global audiences.
  • Timestamps & Multi-language Support: Generate timestamps and handle many languages for captions.

Quick Start

Transcribe a given audio URL into text using the inference.sh CLI.

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 timestamps for podcasts and lectures?

Transcribe audio to text with timestamps by processing the audio through Whisper models via the inference.sh CLI. This generates accurate, time-coded transcripts suitable for podcasts, lectures, and multimedia captioning.

Can I use Whisper models for multilingual transcription and translation?

Whisper models support multilingual transcription and translation through the inference.sh CLI. They convert spoken audio into text across multiple languages and translate transcripts for global audiences.

What is the best way to generate captions from audio for meetings?

Generate captions from meeting audio by running the audio files through Whisper models using the inference.sh CLI. This produces accurate text transcripts with timestamps, enabling accessibility and searchability.

Does this speech-to-text approach work without requiring additional dependencies?

Yes, this speech-to-text approach works without requiring additional dependencies. It uses predefined Whisper models accessible via the inference.sh CLI to directly transcribe audio URLs into text.

Why use Whisper models for transcribing multimedia content instead of other tools?

Whisper models provide built-in multi-language support, timestamp generation, and translation capabilities for multimedia content. They run directly via the inference.sh CLI, streamlining the transcription workflow without external dependencies.

How do I start transcribing a given audio URL into text?

Transcribe an audio URL into text by passing it to the inference.sh CLI. The predefined Whisper models process the audio input and output a searchable text transcript.