Lightning v3.1 Pro

• Overall - Holistic listener rating of how natural the voice sounds end-to-end.
• Naturalness - How human-like the voice sounds; penalizes robotic or synthetic quality.
• Intonation - Whether pitch rises and falls appropriately for the sentence type (question, statement, exclamation).
• Prosody - The broader umbrella of rhythm, stress, and melody, how well the voice “reads” the sentence as a human would.

• Overall - Holistic listener rating of how expressive the voice sounds given the context of the sentence.
• Paralinguistics - Non-verbal vocal elements like laughter, sighs, or filler sounds (“um”, “uh”) and whether they’re rendered appropriately.
• Emotions - How accurately the voice conveys the intended emotional tone (neutral, warm, urgent, etc.).

• Boundary Consistency - Whether phrase and sentence boundaries are marked consistently with pauses or pitch shifts, without arbitrary breaks mid-phrase.
• Pronunciation Style - Not just correctness, but stylistic choices i.e., formal vs. casual register, regional accent consistency, honorific handling.
• Natural Pace - Whether the speaking rate feels comfortable and appropriate for the content type, neither rushed nor dragging.
• Pause Placement - Whether silences appear at semantically correct points (after commas, between clauses) rather than mid-word or mid-phrase.
• Breathing Naturalness - Whether breath sounds occur at realistic points and with realistic frequency, not absent entirely or inserted randomly.

• WER (Word Error Rate) - Percentage of words in the transcript that differ from the reference; measures how faithfully the TTS renders the input text.
• CER (Character Error Rate) - Like WER but at the character level.
• Hallucination - Words or sounds the TTS generates that have no basis in the input text. Insertions, substitutions, or fabricated content.
• Deletion - Words from the reference text that the TTS dropped entirely.
• Pronunciation % - The proportion of words pronounced correctly out of total words.
• Whisper jiwer vs Whisper LLM - Two judging methodologies. jiwer uses raw Whisper-decoded transcripts; LLM-judged uses a follow-on LLM to normalize transcription noise. Both report the same metric family; LLM-judged tends to give lower error rates by reducing false positives from punctuation/casing.

• Emergent Win % - How often listeners preferred Lightning v3.1 Pro over MiniMax speech-2.8-hd when hearing both say the same thing. Above 50 = Lightning won more often than not. Scores averaged over a male and a female voice per language.
• Acc-LLM Advantage - How many fewer word mistakes per hundred words compared to MiniMax, judged with a language-aware LLM. Positive = ours is more accurate.
• Acc-jiwer Advantage - Same as Acc-LLM, but the word mistakes are counted automatically by software (jiwer) rather than an LLM. Positive = ours is more accurate.
• WV-MOS Advantage - Difference in overall sound quality, rated automatically by a quality-prediction model (WV-MOS). Positive = ours sounds cleaner.
• Audiobox-CE Advantage - Difference in how pleasant and enjoyable the audio is to listen to (Content Enjoyment score). This score matches human opinions best in our testing. Positive = ours is more enjoyable.
• Audiobox-PQ Advantage - Difference in recording/production quality (e.g. how clean and professional the audio sounds). Positive = ours is cleaner.

• Mean MOS - Mean Opinion Score: average listener rating on a 1–5 scale across the test set; the canonical aggregate quality metric in TTS evaluation.
• UTMOS - A predicted MOS from the UTMOS reference model - an automated proxy for subjective quality.
• WV-MOS - A predicted MOS from the WavLM-based WV-MOS reference model - another automated proxy commonly reported alongside UTMOS for cross-validation.

Pro runs on dedicated inference capacity, isolated from general Lightning traffic, and ships a curated premium voice catalog with improved naturalness. Concurrency, latency, and rate-limit ceilings are identical to standard Lightning v3.1 - the difference is voice quality and the catalog.

Set model to lightning_v3.1_pro explicitly. The field is optional, but it defaults to standard Lightning v3.1 - for Pro you must pass it on every request.

Pair Pro voice IDs with "model": "lightning_v3.1_pro". The API does not currently reject mismatched pairings, but pairing a Pro voice with "model": "lightning_v3.1" (or omitting model) can produce wrong or hallucinated audio. Server-side validation is on the roadmap.

Yes. Pass model: lightning-v3.1-pro to the voice-cloning API to clone onto the Pro pool, then use the resulting voice_id with TTS model: lightning_v3.1_pro. The default is lightning-v3.1.

Set language: en for UK + American accented English, or language: hi for Indian accented English + Hindi with native code-switching. Omitting language defaults to en + hi.

Pro shares the Lightning v3.1 base-queue voices for English + Hindi (meher, devansh, kartik, maithili, liam, avery), so word events emit on those. Other Pro-only voices fall back to silent graceful degradation - audio is normal, but no word events are emitted.

It is a new Lightning v3.1 Pro checkpoint for English (en), Spanish with English code-switching (es) and Hindi with English code-switching (hi), in preview until 2026-11-28. Try it by passing "model": "lightning_v3.1_pro_10_26" with one of those language values. Every Pro voice works on it; the recommended voices give the best results. On 2026-11-28, it becomes the stable lightning_v3.1_pro for English, Spanish and Hindi only, including requests that omit language, so use "model": "lightning_v3.1_pro" from then on. lightning_v3.1_pro_10_26 remains available for 30 more days, until 2026-12-28. See Checkpoints.

Only if you call lightning_v3.1_pro with language set to en, es or hi, or without language. On 2026-11-28, the lightning_v3.1_pro_10_26 preview becomes the stable lightning_v3.1_pro for English, Spanish and Hindi, and those requests start using it, so test lightning_v3.1_pro_10_26 before then. You don’t need to change the model value. If you are already calling lightning_v3.1_pro_10_26, switch to lightning_v3.1_pro by 2026-12-28; the preview ID remains available for 30 days after promotion. Enterprise customers who need more time can contact their account team to pin their current checkpoint.

No. Only English, Spanish and Hindi move to the new checkpoint. All other languages and auto stay on their current checkpoints, and no action is needed.