Hamsa Media · API
Media Platform API
The same engine behind the media studio, as a jobs API: submit audio for transcription, text for speech, and transcripts for AI-generated content — then list, inspect and manage every job from your backend.
- Transcription jobs
- Text-to-speech jobs
- AI content from transcripts
- Webhook delivery
- Full job management
Capabilities
The studio’s workloads, as API jobs
Transcription by URL
Point the transcribe endpoint at a hosted audio or video file and get back a timestamped, speaker-separated transcript — with SRT subtitles if you ask.
Speech generation
The text-to-speech jobs endpoint turns scripts into audio in the voice and dialect you pick, including voices you cloned.
AI content from transcripts
Generate structured content from completed transcriptions — with ready templates or your own custom prompt — through the AI content endpoints.
Costs known up front
The estimate endpoint prices an AI content job before you run it, so pipelines can decide before they spend.
Jobs you can manage
List jobs, view one with its results, update it or delete it — the job store behind the studio is fully exposed.
Results pushed to you
Jobs deliver to your webhookUrl with optional webhook authentication, or you poll by job ID — your pipeline picks the pattern.
Integration
A media pipeline, request by request
How recorded content moves through the API.
- 01
Create a secret key
Create a secret key in the dashboard and keep it on your server. Every request carries it in the Authorization header as Token <key>.
- 02
Submit the media
POST the file’s URL to the transcribe endpoint with a model and language. The job queues and runs asynchronously.
- 03
Receive the transcript
The completed transcription lands on your webhook, or you poll the job — timestamped segments, speakers, and SRT if requested.
- 04
Generate what comes next
Feed the transcription to the AI content endpoints for summaries and structured documents, or send a script to the TTS jobs endpoint for narrated audio.
- 05
Keep the job store clean
List what has run, inspect results by ID, update titles and metadata, and delete jobs your pipeline no longer needs.
Reference
The media platform API at a glance
Each row links to its page in the API docs — the authoritative reference, with full schemas and runnable code samples.
| Area | Detail |
|---|---|
| Transcription | A hosted media file in, a timestamped transcript out — with SRT subtitles if requested |
| Text to speech | Scripts to audio in the voice and dialect you choose, delivered async |
| AI content | Structured content from transcripts, with ready templates or your own custom prompt |
| Cost estimate | Price an AI content job before you run it |
| From transcripts | Generate content directly off a completed transcription |
| Job management | List jobs, view one with its results, update it or delete it |
| Delivery | Webhook with optional authentication per job, or polling by job ID |
| Models | STT: Hamsa-General-V2.0 and Hamsa-Conversational-V1.0; TTS voices from the catalog or cloned |
| Authentication | A secret key in the Authorization header of every request, as Token <key> |
Use cases
Pipelines teams run on the media API
Publish-time transcription
New episodes and videos are transcribed as they are published, and the transcript lands in your CMS beside the media.
Subtitle automation
Every video that enters your library leaves with tuned SRT subtitles, without an editor opening a tool.
Content repurposing
A recorded webinar becomes a summary, an article and social posts through the AI content endpoints — from one transcription job.
Voice-over generation
Scripts from your localization flow go in, narrated Arabic audio in the right dialect comes out, delivered to your storage by webhook.
FAQ
Questions about the media platform API
How does the API relate to the media studio?
+
Same engine, two surfaces. The studio is the web workspace for people; the API runs the same transcription, speech and AI content workloads headless, from your own backend, with jobs as the shared currency.
How do results get delivered?
+
Each job accepts a webhookUrl, with optional webhook authentication, and posts the result there when it completes. Prefer pulling? Poll the job by its ID instead.
What can the AI content endpoints produce?
+
Structured written content generated from a completed transcription — using ready templates or a custom prompt you pass to the /custom endpoint. The /estimate endpoint prices the job before you run it.
Can I use cloned voices in TTS jobs?
+
Yes. Create the voice through the voice cloning API, preload it once, and use its ID as the voiceId on text-to-speech jobs.
How is the API authenticated?
+
With a secret key created in the dashboard, sent in the Authorization header as Token <key>. It is shown once and refused from a browser, so keep it on your server. Keys are project-scoped, so jobs land in the project the key belongs to.
Keep reading
Hamsa API · Speech to Text
Arabic Speech to Text API
Batch, synchronous and WebSocket transcription of Arabic across dialects, with word timestamps, diarization and SRT export.
Read →Hamsa API · Text to Speech
Arabic Text to Speech API
Synchronous, async and streaming Arabic speech generation in 14 dialects plus English, with telephony output and cloned voices.
Read →Hamsa API · Voice Cloning
Voice Cloning API
Custom Arabic and English voices from your audio samples, created, previewed and managed over REST, usable across TTS, agents and media.
Read →


