feat(audio): add voice note transcription and Islamic audio analysis #140

[Wiseman52]

Overview

This PR adds an offline-first Voice Note Transcription and Islamic Audio Analysis system: upload validation and noise assessment, a deterministic Islamic-audio analysis pipeline over transcribed speech (language/dialect identification, recitation detection, question extraction, terminology recognition, speaker estimate, timeline, tone), and pluggable Gemini/Whisper transcription backends with a responder seam for grounded answer drafting.

Related Issue

Closes #140

Changes

🎙️ Voice-Note Transcription & Islamic Audio Analysis

• [ADD] audio_analysis.py
  • Magic-byte upload validation (WAV/OGG/MP3/M4A) with a 25 MiB cap, plus a WAV noise/SNR assessment that recommends denoising profiles.
  • Offline analysis pipeline: language & Arabic-dialect identification (Egyptian, Levantine, Gulf, Maghrebi vs MSA), Quranic recitation detection, timestamped question extraction, ~60-term Islamic-terminology recognition, a conservative speaker estimate, a timeline of key moments, and an emotional/spiritual tone read via the existing /sentiment engine.
  • Transcriber seam — Gemini (default via GEMINI_API_KEY), OpenAI-compatible Whisper (WHISPER_API_BASE), and offline Static/Passthrough — with streaming chunking for long clips, and a Responder seam for grounded answer drafting.
  • Endpoints: POST /audio/transcribe, POST /audio/analyze, POST /audio/generate, GET /audio/terminology, GET /audio/formats.
• [MODIFY] main.py — register the audio-analysis router.
• [MODIFY] README.md — document the feature, endpoints, and backend selection.
• [ADD] tests/test_audio_analysis.py — 33 offline tests (no network, no keys).
• [MODIFY] .github/workflows/ci.yml — add an audio-analysis test step.

Verification Results

ruff check .                                    ✅ All checks passed
ruff format --check .                           ✅ 187 files already formatted
mypy .                                          ✅ Success: no issues found in 187 source files
pytest -q tests/test_audio_analysis.py          ✅ 33 passed

Acceptance Criteria

Acceptance Criteria	Status
Voice notes are transcribed into timestamped segments	✅ Gemini/Whisper backends plus offline Static/Passthrough fallbacks
Uploads are validated; oversize/mismatched rejected	✅ Magic-byte sniffing, 25 MiB cap, 415 on unrecognised formats
Transcripts analysed offline for Islamic audio context	✅ Language/dialect, recitation, questions, terminology, speakers, timeline, tone
Recognised terminology is browsable	✅ GET /audio/terminology (~60 terms)
CI covers the new module	✅ Dedicated pytest step; full lint/mypy/format suite green

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 4d4843df-8dcd-4096-a056-aedc5fa8f1bb

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[drips-wave]

@Wiseman52 Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits

[zeemscript]

@Wiseman52 this PR has merge conflicts with the main branch. Please resolve the conflicts (merge main in or rebase) and push the fix so it can be merged. Thanks!