Remove 'Apply LLMs' guide and add speech_models param to guides 5-30, 32-34

[devin-ai-integration]

Summary

Two changes in this PR:

1. Remove "Apply LLMs to pre-recorded audio" guide from the Pre-recorded STT → Guides sidebar in fern/docs.yml. The underlying page (pages/lemur/apply-llms-to-audio-files.mdx) is not deleted — it remains accessible via the LLM Gateway tab at /docs/llm-gateway/apply-llms-to-audio-files.

2. Add the now-required speech_models parameter to all transcription code examples across 23 guide files (guides 5–30 and 32–34), with format adjusted per language/SDK:

   • Python SDK: speech_models=["universal-3-pro", "universal-2"] added to TranscriptionConfig or passed as config to transcriber.transcribe()
   • JavaScript SDK: speech_models: ["universal-3-pro", "universal-2"] added to client.transcripts.transcribe() options object
   • REST API / raw requests: "speech_models": ["universal-3-pro", "universal-2"] added to JSON payload

Guides intentionally skipped (no changes needed):

Guide	Reason
13 – Iterate over Speaker Labels with Make.com	UI-based tutorial, no code snippets
16 – Generate Custom Speaker Labels with Pyannote	Already has speech_models
17 – Use Speaker Diarization with Async Chunking	No transcript creation calls (only polls existing transcript IDs)
19 – Use Automatic Language Detection as a Separate Step	Already has speech_models
26 – Troubleshoot Common Errors	Error reference guide, no transcription creation code
31 – Audio File Downsampling Recommendations	No transcription code

Review & Testing Checklist for Human

• Spot-check code examples render correctly in the docs preview after deploy — especially guides with multiple code blocks (e.g., YouTube transcripts, GitHub files, batch transcription)
• Verify the JS SDK change in transcribing-github-files.mdx — the example was changed from positional string transcribe("url") to object form transcribe({ audio_url: "url", speech_models: [...] }) to accommodate the new param. Confirm this matches the JS SDK's API.
• Confirm skipped guides (13, 16, 17, 19, 26, 31) truly don't need speech_models added
• Verify /docs/guides/apply-llms-to-audio-files doesn't 404 after deployment — a redirect may be needed if external links point to the old path
• Confirm the Pre-recorded STT sidebar no longer shows the "Apply LLMs" guide after deploy

Notes

• There are existing redirects in docs.yml (e.g. /docs/getting-started/apply-llms-to-audio-files → /docs/llm-gateway/apply-llms-to-audio-files), but none specifically for /docs/guides/apply-llms-to-audio-files. Consider adding one if that URL has been shared externally.
• In batch_transcription.mdx, the quickstart defines config at module scope while the step-by-step section defines it inside the function — this mirrors the existing pattern for transcriber in the original code.

Link to Devin session: https://app.devin.ai/sessions/a9dc0352c32a4ba4ab2fe32704332411
Requested by: @MAsuamah

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring

[github-actions]

🌿 Preview your docs: https://assemblyai-preview-50f2700e-b2cd-4eb5-bffb-a98aa0bf7697.docs.buildwithfern.com/docs

[github-actions]

🌿 Preview your docs: https://assemblyai-preview-325f8181-da4d-4613-b129-1752290a9deb.docs.buildwithfern.com/docs

[github-actions]

🌿 Preview your docs: https://assemblyai-preview-b91112a6-08dc-48ee-9dc6-3acc77f320b8.docs.buildwithfern.com/docs