Universal Translator
Patient education, CME, research, and ministry video that meets City of Hope's clinical bar and reaches every patient community you serve — out of the box, on every property already running in your Brightcove account.
In a hurry? Listen to a quick overview of this page.
One source video — every patient community, on day one.
Every video City of Hope produces — patient education on Empower My Plan, physician CME on cme.cityofhope.org, research updates from the Stephenson Pancreatic Cancer Institute, donor briefings, and Our Journey of Hope ministry training — needs to land in the viewer's language to drive comprehension, adherence, and trust.
Universal Translator generates captions and speaker-matched dubbed audio in 180+ languages directly inside the Brightcove account you already use — collapsing multi-vendor localization into a single native step, so a Spanish, Mandarin, Korean, or Arabic version ships the same day as the English source.
Why City of Hope needs the Universal Translator.
City of Hope serves patients across California, Arizona, Illinois, and Georgia — communities that speak Spanish, Mandarin, Cantonese, Korean, Tagalog, Vietnamese, Arabic, and more. Today, translating a patient education module or a CME lecture means a separate agency, a separate re-upload, and weeks of turnaround. Universal Translator collapses that into a single native step inside the Brightcove account already powering cityofhope.org, cancercenter.com, and Empower My Plan.
Video Showcase
A live look at the Universal Translator output on a real Brightcove asset.
Celebrating 50 Years of Hope: City of Hope BMT Anniversary Event
City of Hope celebrates 50 years of its life-saving Bone Marrow Transplant program with special guest Will Ferrell.
Captions & Overdubs in 3 Languages
AI-generated, human-reviewable captions and synthesized overdubs covering 3 target languages — English, Spanish, and Chinese — to reach every patient community you serve.
Audio Descriptions in 3 Languages
Narrated audio descriptions ensure visually impaired viewers receive the full context of on-screen visuals, charts, and brand storytelling alongside the primary audio track.
Cost Analysis
| Step | Rate | Calculation | Credits | Cost |
|---|---|---|---|---|
| 1 — Auto-Captions (EN source) | 1 / min | ceil(1.50) = 2 | 2 | $0.12 |
| 2 — Captions Translation (2 langs) | 19 / min / lang | 19 × 1.5026 × 2 | 57.1 | $3.43 |
| 3 — Audio Dubbing (2 langs) | 18 / min / lang | 18 × 1.5026 × 2 | 54.1 | $3.25 |
| Total — Brightcove AI Suite | 113.2 | $6.79 | ||
* Auto-Captions rounds up: 1.50 min → 2 credits. Translation and dubbing use full precision (1.5026 min). Brightcove bills on duration only, so multi-speaker handling adds no extra cost.
Brightcove vs. Third-Party AI Provider
Brightcove AI Suite vs. blended industry AI rates ($12–$16 per language) across HeyGen, ElevenLabs, Rask.ai, and Papercup for a 1.50-min patient celebration event — including captions + audio dubbing across 2 translated languages (es-US, zh).
- · Credits used: 113.2
- · Per language (2 translated): $3.40
- · Per language (3 incl. EN): $2.26
- · Platform overhead: $0
- · Integration required: None
Full pipeline · 3 unique languages (1 EN source + es-US + zh) · captions + audio overdubs.
- · Per language (low): $12.00
- · Per language (high): $16.00
- · Subscriptions: +$5–$330/mo
- · Integration overhead: Yes
- · Captions included: Sometimes
$12/lang × 2 translated languages. Blended AI-tier vendors across the market.
- · Per-language rate: $16.00
- · HeyGen API: ~$27
- · ElevenLabs Pro: ~$26+
- · Captions included: Rarely
- · Medical term handling: Risk of error
$16/lang × 2 translated languages. Top-tier vendors with premium voice quality.
Brightcove vs. Human Labor
Modeled @ $25/hr for a 1.50-min patient celebration event (~120-word emotionally sensitive script): ~1 hr captioning (live event audio, medical neologisms like "transplantiversary") + ~20 min/lang translation + ~1.5 hrs/lang voice recording (patient-facing emotional content) = 4.7 hrs for the conservative model. The realistic w/ QA model layers professional QA review (2× multiplier) and patient-content sensitivity oversight on top of the conservative baseline.
- · Per language: $3.40
- · Turnaround: Minutes
- · Medical neologisms handled: Both langs
- · Human specialists needed: None
- · Emotional tone preserved: Both langs
113.2 credits · minutes not weeks · all 2 translated languages.
- · EN captioning (1 hr): $25
- · Translation (2 × 20 min): $17
- · Voice recording (2 × 1.5 hr): $75
- · Per language total: $57.08
- · Turnaround: Days–Weeks
4.7 hrs × $25/hr. Conservative model — professional voiceover talent typically charges $150–$400/hr per language.
- · QA / emotional review: 2× multiplier
- · Patient content sensitivity oversight
- · Per language total: $114+
- · Captioning + translation + dub
- · Turnaround: Weeks
9.3 hrs × $25/hr. Realistic model layering professional QA review and patient-content sensitivity oversight on top of the conservative baseline.
Summary — all scenarios
Complete captions + overdubs pipeline · 1.50-min video · 3 languages (1 EN source + es-US + zh).
Brightcove credit rates per AI Suite Multiplier Table ver. E.04.01.2026: Auto-Captioning 1 credit/min (rounded up); Captions Translation 19 credits/min/lang; Audio Dubbing 18 credits/min/lang; credit price $0.06. Video duration 90.155 sec (1.5026 min) pulled live from Brightcove Player API on video ID 6398156222112 (account 6302584978001 · short URL bcove.video/4e2TG7h). 3 caption tracks present: 1 English source + 2 translated languages (es-US, zh), all evaluated. Industry AI comparison uses 2 as the billable language count. Brightcove's flat per-minute model includes all supported languages at the same rate, and a multi-speaker video costs the same as a single-speaker video — Brightcove bills on duration only. Conservative human-labor model: 1 hr captioning + 2 × (20 min/lang translation + 1.5 hr/lang voice recording) = 4.7 hrs at $25/hr; professional voiceover talent typically charges $150–$400/hr per language, so this is a conservative floor.
Translation Accuracy Report
Quality benchmark of the Brightcove Universal Translator on the "City of Hope BMT 50th Anniversary Celebration with Will Ferrell" patient celebration event, scored across both translated languages against a four-criteria semantic-fidelity rubric tuned for patient-facing medical content. City of Hope, BMT, Dr. Foreman, patient names (Lynn, Patrick), and the 50-year milestone preserved correctly in every language; the medical neologism "transplantiversary" rendered naturally in both Spanish and Chinese; the warm, intimate register of the celebration preserved end-to-end. The English source is clean — no ASR errors detected — and both translated tracks are deployment-ready.
Both Spanish (es-US) and Chinese (zh) scored 5/5 across all four criteria — translation fidelity, brand & terms, fluency & tone, and completeness. City of Hope, BMT, Dr. Foreman, Lynn, Patrick, and the 50-year milestone preserved correctly. "Transplantiversary" rendered as "aniversarios de trasplante" in Spanish and natural Simplified Chinese — both warm and dignified. The patient testimonial "you gave me hope, you gave me more life, you gave me more time" carries full emotional resonance in both languages, and the Chinese track chose "新生" (new life/rebirth) over a literal rendering — a genuinely better choice for a transplant patient. The English source is clean — no ASR errors — and both translated tracks are deployment-ready.
How we evaluate
Brightcove Universal Translator generates WebVTT caption tracks for each target language directly from the English source. Each track is fetched from the delivery CDN and aligned segment-by-segment against the source transcript, then scored 1 – 5 across four criteria and averaged into a composite accuracy score.
Generate
Universal Translator produces VTT captions for the target language set from the English source.
Align
Each translated caption segment is paired with its English counterpart for direct comparison.
Score
A semantic-fidelity + fluency rubric scores every language 1 – 5 on four criteria.
Translation fidelity
Did the AI correctly render the source meaning end-to-end? The 50-year milestone, patient names, the cells→hope→life→time testimonial parallelism, and the "cancer is the opponent / City of Hope, the victory" rhetorical arc must be exact in every language.
Brand & terminology
City of Hope, BMT (Bone Marrow Transplant), "transplantiversary," and patient names (Lynn, Patrick) must be preserved verbatim. "50 years" milestone must be exact. The emotional register (hope, miracle, life, time) must be conveyed with appropriate warmth.
Fluency & tone
Patient celebration event — emotional, personal, celebratory. Translations should preserve the warmth and intimacy of the original speakers. This is patient-facing content; tone matters as much as accuracy.
Completeness
All speaker segments present. 50-year milestone acknowledged. Patient names on stage (Lynn, Patrick) present. Personal patient testimonial ("you gave me hope, you gave me more life, you gave me more time") complete.
Language-by-language results
3 tracks · 1 EN source + 2 translated languages evaluated
Score scale
- 5/5 — Excellent. Native-quality.
- 4/5 — Minor stylistic issues only.
- 3/5 — Acceptable but non-standard terminology in places.
- 2/5 — Meaning preserved but noticeable errors.
- 1/5 — Steps lost, reordered, or hallucinated.
Tier thresholds
- 90 – 100% — Production-ready.
- 75 – 89% — Light human review recommended.
- < 75% — Native review required.