How mrtds224/alora works
ALORA is a collaborative corpus-collection platform for under-resourced African languages. Volunteers or field collectors provide speaker/session details and record prompt-guided speech; the backend stores audio and metadata, analyzes recording quality, supports review and versioned transcription, and exports a speaker-disjoint corpus. The graph includes documented offline capture and acoustic-QA capabilities. Some subsystem paths and links rely on README or source-index references; unsampled implementations are not treated as inspected.
Main entry points: main.py ALORA server · web.py Web pages · api_public.py Public API · auth.py Authentication and roles
Components 27
Participation and web
- Capture interface
static/js/contribute.js - Audio recorder
static/js/recorder.js - Offline queue
static/js/offline_queue.js - Web pages
app/routers/web.py - Public API
app/routers/api_public.py
Corpus collection
- Session API
app/routers/api_sessions.py - Speaker API
app/routers/api_speakers.py - Prompt API
app/routers/api_prompts.py - Recording API
app/routers/api_recordings.py - Session management
app/services/session.py - Speaker management
app/services/speaker.py - Prompt rotation
app/services/prompt.py - Audio processing
app/services/recording.py - Acoustic QA
app/services/audio_qa.py
Review and annotation
- Quality review
app/services/quality.py - Segmentation API
app/routers/api_segmentation.py - Speech segmentation
app/services/segmentation.py - Transcription API
app/routers/api_transcriptions.py - Versioned transcription
app/services/transcription.py - Export API
app/routers/api_export.py - Corpus export
app/services/export.py
Platform services
- Authentication and roles
app/services/auth.py - Object storage
app/services/storage.py - PostgreSQL
app/core/database.py - ALORA server
app/main.py
Other components
- Volunteer
- Field and review team
Connections 46
- Volunteer → Capture interface: uses
- Field and review team → Web pages: uses
- Capture interface → Audio recorder: records with
- Capture interface → Offline queue: queues offline
- Capture interface → Public API: submits to
- Web pages → Session management: calls
app/routers/web.py - Web pages → Speaker management: calls
app/routers/web.py - Web pages → Prompt rotation: calls
app/routers/web.py - Web pages → Audio processing: calls
app/routers/web.py - Web pages → Quality review: calls
app/routers/web.py - Web pages → Versioned transcription: calls
app/routers/web.py - Public API → Prompt rotation: selects prompts
app/routers/api_public.py - Public API → Session management: manages sessions
app/routers/api_public.py - Public API → Speaker management: manages speakers
app/routers/api_public.py - Public API → Audio processing: submits audio
app/routers/api_public.py - Session API → Session management: calls
app/routers/api_sessions.py - Speaker API → Speaker management: calls
app/routers/api_speakers.py - Prompt API → Prompt rotation: calls
app/routers/api_prompts.py - Recording API → Audio processing: calls
app/routers/api_recordings.py - Recording API → Quality review: reviews quality
app/routers/api_recordings.py - Audio processing → Acoustic QA: analyzes audio
app/services/recording.py - Audio processing → Object storage: stores audio
app/services/recording.py - Segmentation API → Speech segmentation: calls
app/routers/api_segmentation.py - Segmentation API → Audio processing: retrieves recordings
app/routers/api_segmentation.py - Transcription API → Versioned transcription: calls
app/routers/api_transcriptions.py - Transcription API → Audio processing: retrieves recordings
app/routers/api_transcriptions.py - Export API → Corpus export: calls
app/routers/api_export.py - Corpus export → Object storage: reads audio
app/services/export.py - Corpus export → Speech segmentation: uses segments
app/services/export.py - Public API → Authentication and roles: checks access
app/routers/api_public.py - Session API → Authentication and roles: checks roles
app/routers/api_sessions.py - Speaker API → Authentication and roles: checks roles
app/routers/api_speakers.py - Prompt API → Authentication and roles: checks roles
app/routers/api_prompts.py - Recording API → Authentication and roles: checks roles
app/routers/api_recordings.py - Segmentation API → Authentication and roles: checks roles
app/routers/api_segmentation.py - Transcription API → Authentication and roles: checks roles
app/routers/api_transcriptions.py - Export API → Authentication and roles: checks roles
app/routers/api_export.py - ALORA server → Public API: registers
app/main.py - ALORA server → Session API: registers
app/main.py - ALORA server → Speaker API: registers
app/main.py - ALORA server → Prompt API: registers
app/main.py - ALORA server → Recording API: registers
app/main.py - ALORA server → Segmentation API: registers
app/main.py - ALORA server → Transcription API: registers
app/main.py - ALORA server → Export API: registers
app/main.py - ALORA server → Web pages: registers
app/main.py