Synthesia

Role: You are a Synthesia video producer converting slides into a single presenter-led video. Constraints: output a 5-minute script for a single avatar, 8-10 scenes mapped to slides, each scene 25-40 seconds; choose an avatar name from Synthesia's library (e.g., 'Ava'), standard neutral English voice, include on-screen headline and one supporting bullet per scene, generate closed captions (SRT). Output format: JSON array 'scenes' with fields: slide_number, start_time, end_time, avatar, voice, speaker_script, on_screen_text, srt_captions. Example scene: {"slide_number":1,"start_time":"00:00:00","end_time":"00:00:30","avatar":"Ava","voice":"en-US-neutral","speaker_script":"Welcome...","on_screen_text":"Course overview","srt_captions":"1\n00:00:00,000 --> 00:00:30,000\nWelcome..."}.

Role: You are writing a 90-120 second CEO update script for Synthesia. Constraints: single avatar (professional, authoritative), tone: concise and optimistic, include exactly three business updates (one metric, one initiative, one team shoutout), one 15-second closing CTA, and provide SRT captions and suggested on-screen headline and lower-third text. Output format: provide a single JSON object with fields: duration_seconds, avatar, voice, full_script, timestamps (start/end for each update), on_screen_elements (headline, lower_third), srt_captions. Example: {"duration_seconds":105,"avatar":"Ethan","voice":"en-GB-formal","full_script":"..."}.

Role: You are a product marketing writer preparing five localized 30-second promo scripts for Synthesia. Constraints: produce one script per locale (US English, UK English, Mexican Spanish, German, French), keep 30±3 seconds each, use the same avatar appearance but choose voice/accent per locale, include localized opening hook, three key product benefits (one sentence each), localized tagline translation, and CTA. Output format: JSON array of 5 objects: {locale, avatar, voice, duration_seconds, script, on_screen_text, translated_tagline}. Example item: {"locale":"es-MX","avatar":"Maya","voice":"es-MX-female","script":"...","translated_tagline":"Tu herramienta, tu ventaja"}.

Role: You are an L&D producer converting 20 slides into four microlearning videos for Synthesia. Constraints: create 4 videos (~3 minutes each), map slide ranges to each video, include scene-level speaker script, one 1-question knowledge check at the end of each video (MCQ with 4 options and correct answer), include captions and suggested thumbnail text, use brand voice (concise, supportive). Output format: JSON with videos array where each video has: video_id, slide_start, slide_end, duration_seconds, scenes[], quiz{question,options,correct_index}, thumbnail_text. Example quiz: {"question":"What's the primary benefit?","options":["A","B","C","D"],"correct_index":2}.

Role: You are a compliance learning designer and Synthesia producer building a 5-part training series. Instructions: produce five 6-8 minute modules covering Policy, Risk, Reporting, Case Studies, and Certification; for each module provide a scene-by-scene script with timestamps, avatar selection (senior neutral presenter), two scenario-based interactive decision points per module with branching text (if trainee selects A -> redirect to remediation scene ID X; if B -> continue), three assessment questions per module with scoring rubric, required captions, and suggested graphics (charts/icons). Output format: JSON {modules: [{id,title,duration,scenes[],branches[],assessments[],srt_captions}]}. Few-shot example: module snippet: {"id":2,"title":"Risk","scenes":[{"scene_id":"2.1","start":"00:00","end":"00:45","script":"..."}],"branches":[{"decision_id":"D1","prompt":"...","options":[{"opt":"A","goto":"remed_2A"},...] }],...}.

Role: You are an enterprise video strategist creating a 7-video onboarding program for Synthesia with compliance and governance steps. Requirements: deliver seven 2-4 minute scripts (welcome, values, IT security, HR policies, product overview, first-90-days, wrap-up), specify custom-avatar usage instructions (consent, legal approval text), localization needs (EN/ES/FR), metadata tags for DAM (title, keywords, retention_policy), access control checklist (who can export/edit), and a production checklist (reviewers, caption QA, final sign-off). Output format: JSON {program:{videos:[],avatar_instructions,localization,metadata_template,access_control,production_checklist}}; include a short example video object.

Compare Synthesia with HeyGen, D-ID, Elai, Colossyan, VEED. Choose based on workflow fit, pricing limits, integrations, governance needs and whether the output must be production-ready or only assistive.