Use case · Long-form video
Narrate long-form videos in Arabic at studio quality.
Long-form rewards the voice that doesn't get tiring at minute twelve. Hakim Faseeh v2 ships studio-grade narration across 15 Arabic dialects with non-verbal tag support and prosody controls that survive a 20-minute script. Batch API renders an hour of audio in minutes; multi-format export drops cleanly into Premiere, DaVinci Resolve, and the LMS uploader your course runs on.
Three steps from script to final cut
Bring your script
Markdown, plain text, or SSML if you want fine-grained prosody control. Hakim accepts scripts up to 100,000 characters in a single batch; long enough for a feature-length explainer or a multi-episode course module.
Voice + dialect, with prosody
Pick a voice from the Hakim Faseeh v2 voice catalogue (filterable by dialect, gender, age, and style; narrator, lecturer, storyteller, news). Apply per-line emphasis, pace, and pause tags inline (the renderer respects the full non-verbal tag set: [laugh], [breath], [whisper], [think], [ask]). Natural prosody, not the robotic literal interpretation lower-tier engines apply.
Render in batch, drop into your editor
Submit the script via the batch endpoint; receive a webhook when it's ready. Per-line audio files plus a master mixdown plus a CSV of timestamps land in your S3 bucket (or Hakim's, your call). Drag onto your timeline, sync to picture, ship.
Built for long-form production
Studio-grade voice quality
Hakim Faseeh v2 was built for long-form. The voices don't develop the synthetic-edge fatigue that lower-tier engines exhibit at minute ten. Listeners stay engaged through a 20-minute course module without consciously noticing the voice is synthesised.
Prosody that survives a script
Per-line
emphasis,pace, andpausetags inline in the script. The renderer applies them with natural intonation rather than literal timing. Punctuation is read; bracketed asides are softer; question marks lift the cadence. The defaults are good; the controls are there when you need them.Batch API + webhooks
Submit a 100k-character script in one POST; receive a webhook when rendering is done. An hour of audio renders in under ten minutes. Per-line files and a master mixdown both ship; pick which one your editor wants.
Multi-format export with timestamps
MP3 for review copies, WAV at 48 kHz for masters, FLAC for archive. Per-line timestamps in CSV so your editor can place lines on the timeline algorithmically. Premiere, DaVinci Resolve, and Final Cut Pro all import the CSV via standard scripts.
Request
Batch render in five lines
curl https://api.tryhakim.ai/v1/tts/batch \
-H "Authorization: Bearer $HAKIM_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "hakim-arab-v2",
"voice_id": "saif-msa",
"script_url": "https://your-bucket.s3.amazonaws.com/episode-12.txt",
"format": "wav",
"sample_rate": 48000,
"webhook_url": "https://your-app.com/hakim/done"
}'Built for these long-form workflows
YouTube long-form creators
Faceless channels, narrated explainers, deep-dive video essays. The hour-of-audio-in-ten-minutes batch turnaround means a creator can ship a 30-minute video on the same day they finish the script. The dialect catalogue lets a single creator address regional sub-audiences without re-recording.
Online courses + LMS content
Coursera, Udemy, locally-hosted LMS platforms. Hakim Faseeh v2 ships course-grade narration across 15 dialects so an instructor can deliver a Saudi-targeted version and an Egyptian-targeted version of the same module without re-recording. CSV timestamps integrate with the LMS chapter-marker import.
Explainer + product videos
B2B product walkthroughs, pitch decks with narration, internal training content. Studio-grade quality without the studio booking. Brand voice cloning available on enterprise so the same voice plays across all marketing video assets.
News + media segments
News organisations producing regional video content. MSA for the broadcast tier; regional dialects for the social-cut tier. Batch API processes a day's worth of segments in a single overnight run.
