The Audio Intelligence API for Ethiopian Languages
Integrate enterprise speech-to-text, speaker diarization, and semantic audio search into your software with simple REST and WebSocket endpoints. Purpose-built for Amharic, Afaan Oromo, and Tigrinya.
45 free minutes on signup · Telebirr & CBE Birr supported · Instant export to DOCX, SRT, VTT & JSON
API Streaming Response Example: POST /v1/transcriptions
JSON Payload (Amharic & English) • REST API Output • 00:45
curl -X POST https://api.andebetai.com/v1/transcribe -H 'Authorization: Bearer ANDEBET_KEY'
“curl -X POST https://api.andebetai.com/v1/transcribe -H 'Authorization: Bearer ANDEBET_KEY'”
{"language":"am","confidence":0.982,"text":"የኢትዮጵያ ንግድ ባንክ የዲጂታል አገልግሎቶች..."}
“{"language":"am","confidence":0.982,"text":"Commercial Bank of Ethiopia digital services..."}”
Built for Scalable Engineering Pipelines
Everything developers need to build production voice interfaces, call center analytics, and qualitative search applications.
Asynchronous Batch & Webhooks
Submit long audio recordings up to 2GB asynchronously and receive secure webhook notifications when transcripts are ready.
Low-Latency WebSocket Streaming
Stream raw PCM audio frames over WebSocket with optimized real-time delivery for live captions and conversational voice bots.
Speaker Diarization Included
Every response includes millisecond start/end timestamps and assigned speaker labels with confidence scores per turn.
Model Context Protocol (MCP)
Natively integrate audio search and transcript retrieval into Claude Desktop and AI agent frameworks using our official MCP server.
How to Get Started with the API in 5 Minutes
From signup to your first audio transcription in under four commands.
Create Free API Key
Sign up and generate your secret API key from the developer portal. Every account includes 45 free minutes.
Instant key provisioning
Install Client Library
Install our official package via npm (npm i @andebet/sdk) or pip (pip install andebet) or use standard cURL requests.
Python & TypeScript SDKs
Submit Audio File
Send audio via HTTP POST with parameters for language selection, diarization, and optional English translation.
Simple 10-line integration
Receive Structured JSON
Consume structured transcripts with word-level timestamps, speaker IDs, and confidence scores in your application.
Standard JSON response
Developer Use Cases
Powering products across fintech, media, education, and enterprise SaaS.
Telecom & Banking IVR Bots
Fintech, banks, telecom providersEnable voice-driven interactive phone menus in Amharic and Afaan Oromo for balance checks and transfer confirmations.
Media Subtitling Pipelines
Broadcasters, video platforms, YouTube automationAutomatically generate synchronized SRT/VTT subtitle files for Horn of Africa video broadcasts.
Call Center Quality Assurance
Support teams, customer care centersTranscribe and analyze thousands of customer service calls to track resolution time and detect user frustration.
AI Agent Tools & Search
AI startups, autonomous agent developersEquip LLM agents with real-time access to audio transcripts and semantic audio retrieval tools.
Frequently Asked Questions: Audio Intelligence API
Pricing, throughput limits, SDKs, and data privacy.
For batch processing via POST /v1/transcribe, a 1-hour audio file is typically processed in under 2 minutes. For live applications, our WebSocket streaming interface is engineered for low-latency delivery from spoken speech to Ge'ez fidel text.
We support WAV, MP3, M4A, AAC, FLAC, OGG, WebM, and MP4 containers. Audio is automatically normalized and resampled by our processing pipeline.
Starter tiers include 10 concurrent requests and 60 requests/minute. Commercial and Enterprise tiers support thousands of concurrent audio streams with a 99.9% uptime SLA.
Yes, 100%. We enforce strict data sovereignty: customer audio is encrypted in transit and at rest, and we never train public models on your API payloads.
Developer Resources & Integrations
Developer Portal & API Docs
Full API endpoint reference, authentication guides, and schemas.
MCPModel Context Protocol (MCP) Server
Connect Andebet directly to Claude Desktop and AI agent tools.
PricingAPI Pricing & Rate Limits
Transparent pricing per audio minute with generous free tiers.
UpdatesAPI Changelog
Latest updates, model releases, and API version notes.
Turn recorded Ethiopian conversations into searchable truth.
Join researchers, media networks, and global organizations extracting structured knowledge from Amharic, Afaan Oromo, and Tigrinya audio.
- 45 free minutes included
- No credit card required
- Instant API & Workspace access
- Enterprise data sovereignty