Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Removed manual installation instructions referencing `references/manual-install.md`. - Updated installation section: now directs users to download binaries from the GitHub releases page for manual installation. - Added a note linking to the Speechall API key creation page in the authentication section. - No functional CLI changes. Documentation only.
What This Skill Does
CLI tool for transcribing audio and video files to text via the Speechall API. Routes requests through multiple speech-to-text providers from a single interface. Supports speaker diarization, subtitle formats (SRT, VTT), and custom vocabulary.
Provides access to speech-to-text models from OpenAI, Deepgram, AssemblyAI, Google, and others through a single CLI without separate SDKs or account integrations for each provider.
When to use it
- Transcribing a recorded interview to a text file
- Generating SRT subtitles for a recorded presentation
- Identifying speakers in a multi-person meeting recording
- Processing domain-specific audio with custom terminology
- Listing available STT models to compare provider options
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Transcribe this team meeting recording with speaker labels and export as SRT subtitles
- 1Set SPEECHALL_API_KEY environment variable with the user's API key
- 2Run speechall meeting.wav --diarization --speakers-expected 3 --output-format srt
- 3Redirect stdout to save the output file: speechall meeting.wav ... > meeting.srt
- 4Confirm the SRT file contains timestamped, speaker-labeled segments
meeting.srt file with timestamped captions and speaker labels for each participant
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.