ElevenLabsAudio modelActive
Scribe v2 Realtime
PopularityRank 481 of 631
29out of 100same as last month
based on two data groups, mostly search
Summary
Scribe v2 Realtime is ElevenLabs’ live speech recognition model. It delivers partial transcripts in about 150 ms according to the provider, covers more than 90 languages, and accepts audio in chunks while returning text continuously. Voice Activity Detection segments speech automatically from periods of silence.
Scribe v2 Realtime supports PCM from 8 to 48 kHz and μ-law, plus manual commit control for finalized segments. Those protocol boundaries determine when an audio chunk is treated as complete. Unlike Scribe v2, it is built for streaming rather than a later batch pass.
All benchmark results
Every published result with its source, test setup and comparison models.
Show all 266 published benchmark results
Page 1 of 23
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
VoiceCodeBench: CTEM (Show measurement, test conditions, and source)
- Source value
- 86.842
- Score
- 86.842
- Metric
- canonical token/entity match
- Unit
- %
- Benchmark version
- 1
- Category
- general
- Task
- ctem
- Task label
- CTEM
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Vals AI
- Status
- active
- Retrieved at
- 2026-11-06
- Methodology
- Source model ID: elevenlabs/scribe_v2_realtime. Source model name: Scribe v2 Realtime. Provider reported by source: Elevenlabs. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-09-24. Latency in seconds: 77.696. Standard error: 0. Cost per task in USD: 0.007263. Source ID: vals-ai. Evidence capture ID: vals-voice-code-bench-2026-11-06-7753d162bfe3. Source payload hash: 7753d162bfe3fa3470d586da3917bf8ddd96f14e10f9aa8f03dd517b8d5f003b. Parser version: vals-astro-v4. Imported at: 2026-11-06. Source cohort: vals-voice-code-bench:1:ctem
VoiceCodeBench: Overall (Show measurement, test conditions, and source)
- Source value
- 55.333
- Score
- 55.333
- Metric
- task success rate
- Unit
- %
- Benchmark version
- 1
- Category
- general
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Vals AI
- Status
- active
- Retrieved at
- 2026-11-06
- Methodology
- Source model ID: elevenlabs/scribe_v2_realtime. Source model name: Scribe v2 Realtime. Provider reported by source: Elevenlabs. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-09-24. Latency in seconds: 77.696. Standard error: 0. Cost per task in USD: 0.007263. Source ID: vals-ai. Evidence capture ID: vals-voice-code-bench-2026-11-06-7753d162bfe3. Source payload hash: 7753d162bfe3fa3470d586da3917bf8ddd96f14e10f9aa8f03dd517b8d5f003b. Parser version: vals-astro-v4. Imported at: 2026-11-06. Source cohort: vals-voice-code-bench:1:overall
Artificial Analysis AA-WER Streaming (final transcription) (Show measurement, test conditions, and source)
- Source value
- 3.5946
- Score
- 3.5946
- Metric
- word error rate (WER)
- Unit
- %
- Benchmark version
- AA-WER v2.2 streaming
- Category
- audio
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Independent evaluation. Direct API streaming evaluation; final-transcription WER across AA-AgentTalk (50%), VoxPopuli (25%), and Earnings22 (25%), about eight hours total. WER is duration-weighted within dataset; Earnings22 chunking is reassembled at parent-call level. AA-WER v2.2 streaming protocol.
Coval pooled corpus-level STT word error rate (Show measurement, test conditions, and source)
- Source value
- 4.333868660541119
- Score
- 4.333868660541119
- Metric
- Pooled corpus-level speech-recognition word error rate, Coval pooled_value
- Unit
- %
- Benchmark version
- Coval rolling 30-day window, re-measured daily; route last measured 2026-10-16
- Category
- audio
- Direction
- Lower is better
- Sample
- 26661
- Source type
- official-leaderboard
- Evaluator
- Coval
- Status
- superseded
- Retrieved at
- 2026-10-16
- Methodology
- Official benchmark leaderboard. Coval endpoint key elevenlabs/scribe_v2_realtime, exact ElevenLabs API model ID scribe_v2_realtime, hosted by ElevenLabs. This is the real-time transcription endpoint, not the separate asynchronous Scribe v2 profile. Rolling 30-day window, re-measured daily.
- Context
- This is Coval's pooled corpus-level STT WER, not the separate per-clip mean_value or a rounded leaderboard headline. Pooled WER and mean per-clip WER are different aggregations and should not be compared as the same score. The WER row's sample_count is used, not another metric's count or a roster total. The result describes Coval's hosted 30-day endpoint cohort, not local or native performance.
Pipecat STT Benchmark semantic WER mean (Show measurement, test conditions, and source)
- Source value
- 3.05
- Score
- 3.05
- Metric
- Arithmetic mean semantic word error rate across successfully transcribed samples
- Unit
- %
- Benchmark version
- Pipecat STT Benchmark at commit 94f567299a21f8d617039bc69f137a131f5dedd5, September 2026 semantic-WER judge update
- Category
- audio
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Pipecat STT Benchmark
- Status
- active
- Retrieved at
- 2026-10-16
- Methodology
- Independent evaluation. The pinned results table reports semantic WER over 1000 attempted samples from pipecat-ai/smart-turn-data-v3.1-train. WER Mean is the average across samples with a returned transcript; the table reports 99.7% transcript success, but does not give the exact number of WER-scored samples. Semantic scoring by Claude Sonnet 5.5 ignores punctuation, contractions, fillers, and equivalent phrasings. The source does not document a language distribution; its dataset preview shows English transcripts.
- Context
- No exact WER-scored N or model-specific confidence interval is published. The 1000 figure is attempted samples, not the WER denominator. The README says differences below about 0.25 percentage points are within the 95% bootstrap interval among the most accurate services, so do not rank close scores.
Pipecat STT Benchmark semantic pooled WER (Show measurement, test conditions, and source)
- Source value
- 3.13
- Score
- 3.13
- Metric
- Pooled semantic word error rate, total judged errors divided by total reference words
- Unit
- %
- Benchmark version
- Pipecat STT Benchmark at commit 94f567299a21f8d617039bc69f137a131f5dedd5, September 2026 semantic-WER judge update
- Category
- audio
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Pipecat STT Benchmark
- Status
- active
- Retrieved at
- 2026-10-16
- Methodology
- Independent evaluation. The pinned results table reports semantic WER over 1000 attempted samples from pipecat-ai/smart-turn-data-v3.1-train. Pooled WER is total judged semantic word errors divided by total reference words, among samples with a returned transcript; the table reports 99.7% transcript success but not the exact scored N. Semantic scoring by Claude Sonnet 5.5 ignores punctuation, contractions, fillers, and equivalent phrasings. The source does not document a language distribution; its dataset preview shows English transcripts.
- Context
- No exact WER-scored N or model-specific confidence interval is published. The 1000 figure is attempted samples, not the WER denominator. The README says differences below about 0.25 percentage points are within the 95% bootstrap interval among the most accurate services, so do not rank close scores.
Coval rolling-30-day pooled word error rate, 2026-10-17 snapshot (Show measurement, test conditions, and source)
- Source value
- 4.3
- Score
- 4.3
- Metric
- Pooled endpoint word error rate (WER)
- Unit
- %
- Benchmark version
- Coval daily fixed-audio endpoint benchmark, rolling 30-day window; route last measured 2026-10-17
- Category
- audio
- Direction
- Lower is better
- Sample
- 26660
- Source type
- independent-evaluator
- Evaluator
- Coval Voice AI Benchmarks
- Status
- superseded
- Retrieved at
- 2026-10-17
- Methodology
- Independent evaluation. Exact ElevenLabs Scribe v2 Realtime streaming endpoint, not the separate asynchronous Scribe v2 model. Coval remeasures fixed public speech audio daily and aggregates over a rolling 30-day window. The 2026-10-17 page reports 4.3% WER on 26660 samples, rounded to one decimal place.
- Context
- This is a new dated rolling-window snapshot. The current page exposes WER only to one decimal place, so it does not supersede the more precise 2026-10-16 value; no confidence interval is shown.
Speko FLEURS read-speech streaming WER (Show measurement, test conditions, and source)
- Source value
- 4.32
- Score
- 4.32
- Metric
- word error rate (WER)
- Unit
- %
- Benchmark version
- Speko STT route elevenlabs:scribe_v2_realtime; cell measured 2026-10-01
- Category
- speech-to-text / streaming
- Direction
- Lower is better
- Sample
- 1541
- Source type
- independent-evaluator
- Evaluator
- Speko speech-to-text benchmark
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Independent evaluation. Exact Speko route elevenlabs:scribe_v2_realtime, displayed model Scribe v2 Realtime; streaming/live socket with audio delivered at real-time pace; FLEURS read-English test+dev+train; live socket at real-time pace. Measured 2026-10-01.
- Context
- No confidence interval is published for this WER cell. The overall board headline retains an older date, while the exact cell-level measured field is 2026-10-01 and is used here. The route is Scribe v2 Realtime, not batch Scribe v2.
Speko EdAcc spontaneous-conversation WER (Show measurement, test conditions, and source)
- Source value
- 12.54
- Score
- 12.54
- Metric
- word error rate (WER)
- Unit
- %
- Benchmark version
- Speko STT route elevenlabs:scribe_v2_realtime; cell measured 2026-10-01
- Category
- speech-to-text / streaming
- Direction
- Lower is better
- Sample
- 283
- Source type
- independent-evaluator
- Evaluator
- Speko speech-to-text benchmark
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Independent evaluation. Exact Speko route elevenlabs:scribe_v2_realtime, displayed model Scribe v2 Realtime; streaming/live socket with audio delivered at real-time pace; EdAcc spontaneous accented conversation; 60 speakers; two realtime passes. Measured 2026-10-01.
- Context
- No confidence interval is published for this WER cell. The overall board headline retains an older date, while the exact cell-level measured field is 2026-10-01 and is used here. The route is Scribe v2 Realtime, not batch Scribe v2.
AssemblyAI normalized WER mean across three noise conditions (Show measurement, test conditions, and source)
- Source value
- 7.78
- Score
- 7.78
- Metric
- normalized word error rate, direct reported mean over no, low and heavy noise
- Unit
- %
- Benchmark version
- AssemblyAI benchmark snapshot; runs 2026-08-27 to 2026-09-22
- Category
- speech-to-text
- Direction
- Lower is better
- Source type
- vendor-reported
- Evaluator
- AssemblyAI speech benchmark
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. AssemblyAI's benchmark page directly reports the exact row Scribe v2 Realtime, with its mean of three conditions, no noise, low noise and heavy noise. The displayed row is 7.78%, with condition values 6.18%, 7.53% and 9.64%. Run window: 2026-08-27 to 2026-09-22; the page states one row per model, most recent run. The bundle records the published mean, not a recomputation.
- Context
- AssemblyAI is the evaluator and also a speech vendor. No per-model N, CI or immutable competitor serving revision is published.
AssemblyAI normalized medical entity error rate mean across three sets (Show measurement, test conditions, and source)
- Source value
- 7
- Score
- 7
- Metric
- normalized medical entity error rate, direct reported mean over three clinical sets
- Unit
- %
- Benchmark version
- AssemblyAI benchmark snapshot; runs 2026-08-18 to 2026-09-22
- Category
- medical speech-to-text
- Direction
- Lower is better
- Source type
- vendor-reported
- Evaluator
- AssemblyAI speech benchmark
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. AssemblyAI's medical-entity table directly reports the exact Scribe v2 Realtime row and the mean over primary care, clinical dictation and respiratory sets. The displayed mean is 7.00%, with 10.89%, 8.01% and 2.09% for those sets. Run window: 2026-08-18 to 2026-09-22; the page states one row per model, most recent run.
- Context
- The row is vendor-published comparative data. No per-model N, CI or immutable competitor serving revision is published.
FLEURS / reported accuracy across 30 European and Asian languages (Show measurement, test conditions, and source)
- Source value
- 93.5
- Score
- 93.5
- Metric
- Publisher-reported accuracy; calculation not specified
- Unit
- %
- Benchmark version
- ElevenLabs Scribe v2 Realtime FLEURS chart, 30-language comparison; evaluation date not stated
- Category
- automatic speech recognition
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- ElevenLabs, Introducing Scribe v2 Realtime
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. The exact source model name is Scribe v2 Realtime and the article says it is available through the ElevenLabs API. Its chart is titled as FLEURS results across 30 European and Asian languages, and the article reports 93.5% accuracy across 30 commonly used languages. The source does not specify the accuracy formula, language list, data split, evaluation date, per-model denominator, API revision, or generation/decoding settings. A separate sentence mentions 500 hard samples but does not explicitly bind that count to this 30-language result.
- Context
- The published accuracy is not interchangeable with word error rate, and no conversion is made. This is a distinct 30-language aggregate, not the English-only FLEURS streaming WER already in the frozen input. The article page states a publication date of 2025-11-11, but does not date the evaluation. No N or confidence interval is published or inferred.
Page 1 of 23
Popularity compared
How often the model is searched for, used and mentioned, compared with every model of its category.
Popularity index
0 to 100, higher is more popular
20 of 45 models
As of November 2026 · Source: gradually.ai popularity index
Over time
Weekly values from October 18, 2026 to November 7, 2026
Scribe v2 RealtimeIndex-Echo-S2ST-2B 30Voxtral Mini TTS 29MOSS-Audio-4B-Thinking 29Index-Echo-S2ST-9B 29
Method changes in the timeline, a jump at these points is no market event:
- 1November 1, 2026: Value only with search or usage data, 30-day attention
What is gaining popularity right now, across models, vendors, harnesses and AI subscriptions, is collected in All AI trends.
Scribe v2 Realtime compared
- 1Scribe v2 Realtime vs. Gemini 3.5 Transcribe Live21 shared tests
- 2Scribe v2 Realtime vs. GPT-Live-Transcribe4 shared tests
- 3Scribe v2 Realtime vs. GPT-Realtime-Whisper3 shared tests
- 4Scribe v2 Realtime vs. Flux19 shared tests
- 5Scribe v2 Realtime vs. Nova-321 shared tests
- 6Scribe v2 Realtime vs. Voxtral Mini Transcribe Realtime6 shared tests
- 7Scribe v2 Realtime vs. Cartesia Ink-23 shared tests
- 8Scribe v2 Realtime vs. Muse Voice Transcribe5 shared tests
Similar models
More models of the same kind as Scribe v2 Realtime, chosen by model line, developer, size, price and release date.
- Eleven v3ElevenLabs · same developer
- Scribe v2ElevenLabs · same developer
- Eleven v3 ConversationalElevenLabs · same developer
- Eleven Flash v2.5ElevenLabs · same developer
- Eleven Multilingual STS v2ElevenLabs · same developer
- Eleven Text to Sound v2ElevenLabs · same developer
- Eleven Music v2.5ElevenLabs · same developer
- Scribe v2 MedicalElevenLabs · same developer
All specifications
Architecture
Input and output
Watermarking and provenance
AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.
- AI watermarking
- Unknown
- no source
- Watermarking technique
- Not known
- no source
Licence and access
API
Further details
- Live transcription
- Over 90 languages; about 150 ms latency according to ElevenLabs
- Source: Live transcription
Sources and data date
Show all 37 sources
Benchmark sources at a glance
Which independent evaluators tested this model. This is coverage, not a quality ranking.
| Evaluator | Test series | Composite indices | Recorded active measurements |
|---|---|---|---|
| Vals AI | 1 | 0 | 3 |
Without vendor claims and historical or disputed results; those stay labelled in the single values.