---name: elevenlabs-sttdescription: ElevenLabs speech-to-text with Scribe models and forced alignment via inference.sh CLI. Models: Scribe v1/v2 (98%+ accuracy, 90+ languages). Capabilities: transcription, speaker diarization, audio event tagging, word-level timestamps, forced alignment, subtitle generation. Use for: meeting transcription, subtitles, podcast transcripts, lip-sync timing, karaoke. Triggers: elevenlabs stt, elevenlabs transcription, scribe, elevenlabs speech to text, forced alignment, word alignment, subtitle timing, diarization, speaker identification, audio event detection, eleven labs transcribeallowed-tools: Bash(belt *)---

Install the belt CLI skill: npx skills add belt-sh/cli

ElevenLabs Speech-to-Text

High-accuracy transcription with Scribe models via inference.sh CLI.

Quick Start

Requires inference.sh CLI (belt). Install instructions

belt login

# Transcribe audio
belt app run elevenlabs/stt --input '{"audio": "https://audio.mp3"}'

Available Models

Model	ID	Best For
Scribe v2	scribe_v2	Latest, highest accuracy (default)
Scribe v1	scribe_v1	Stable, proven

• 98%+ transcription accuracy
• 90+ languages with auto-detection

Examples

Basic Transcription

belt app run elevenlabs/stt --input '{"audio": "https://meeting-recording.mp3"}'

With Speaker Identification

belt app run elevenlabs/stt --input '{
  "audio": "https://meeting.mp3",
  "diarize": true
}'

Audio Event Tagging

Detect laughter, applause, music, and other non-speech events:

belt app run elevenlabs/stt --input '{
  "audio": "https://podcast.mp3",
  "tag_audio_events": true
}'

Specify Language

belt app run elevenlabs/stt --input '{
  "audio": "https://spanish-audio.mp3",
  "language_code": "spa"
}'

Full Options

belt app run elevenlabs/stt --input '{
  "audio": "https://conference.mp3",
  "model": "scribe_v2",
  "diarize": true,
  "tag_audio_events": true,
  "language_code": "eng"
}'

Forced Alignment

Get precise word-level and character-level timestamps by aligning known text to audio. Useful for subtitles, lip-sync, and karaoke.

belt app run elevenlabs/forced-alignment --input '{
  "audio": "https://narration.mp3",
  "text": "This is the exact text spoken in the audio file."
}'

Output Format

{
  "words": [
    {"text": "This", "start": 0.0, "end": 0.3},
    {"text": "is", "start": 0.35, "end": 0.5},
    {"text": "the", "start": 0.55, "end": 0.65}
  ],
  "text": "This is the exact text spoken in the audio file."
}

Forced Alignment Use Cases

• Subtitles: Precise timing for video captions
• Lip-sync: Align audio to animated characters
• Karaoke: Word-by-word timing for lyrics
• Accessibility: Synchronized transcripts

Workflow: Video Subtitles

# 1. Transcribe video audio
belt app run elevenlabs/stt --input '{
  "audio": "https://video.mp4",
  "diarize": true
}' > transcript.json

# 2. Build caption segments (start, end, text) from the word timestamps in step 1
belt app run infsh/caption-videos --input '{
  "video_file": "https://video.mp4",
  "segments": [{"start": 0.0, "end": 2.5, "text": "<text-from-step-1>"}]
}'

Supported Languages

90+ languages including: English, Spanish, French, German, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, Hindi, Russian, Turkish, Dutch, Swedish, and many more. Leave language_code empty for automatic detection.

Use Cases

• Meetings: Transcribe recordings with speaker identification
• Podcasts: Generate transcripts with audio event tags
• Subtitles: Create timed captions for videos
• Research: Interview transcription with diarization
• Accessibility: Make audio content searchable and accessible
• Lip-sync: Forced alignment for animation timing

Related Skills

# ElevenLabs TTS (reverse direction)
npx skills add inference-sh/skills@elevenlabs-tts

# ElevenLabs dubbing (translate audio)
npx skills add inference-sh/skills@elevenlabs-dubbing

# Other STT models (Whisper)
npx skills add inference-sh/skills@speech-to-text

# Full platform skill (all apps)
npx skills add inference-sh/skills@infsh-cli

Browse all audio apps: belt app list --category audio