CartesiaAudio modelActive
Sonic 3.6
PopularityRank 279 of 875
39out of 100same as last month
Summary
Sonic 3.6 is Cartesia’s fast, natural-sounding text-to-speech model with support for 44 languages. It follows the transcript closely, speaks confirmation codes and difficult heteronyms without preprocessing, and adapts pauses and pacing to context. The documented language expansion also includes Latin-script Hindi, Odia, and Urdu.
Cartesia uses sonic-3.6 as an alias for the latest stable snapshot, while sonic-3.6-2026-08-27 stays on that dated version. The two identifiers therefore differ in reproducibility, not in the documented model family.
All benchmark results
Every published result with its source, test setup and comparison models.
Show all 147 published benchmark results
Page 1 of 13
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.
Coval pooled corpus-level synthesized-speech word error rate (TTS WER) (Show measurement, test conditions, and source)
- Source value
- 1.4981466606730975
- Score
- 1.4981466606730975
- Metric
- Pooled corpus-level synthesized-speech 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
- 2100
- Source type
- official-leaderboard
- Evaluator
- Coval
- Status
- superseded
- Retrieved at
- 2026-10-16
- Methodology
- Official benchmark leaderboard. Coval endpoint key cartesia/sonic-3.6, exact Cartesia API model ID sonic-3.6, hosted by Cartesia. This is a model-scoped synthesized-speech WER observation; the public route does not identify the tested voice. It is not STT WER. Rolling 30-day window, re-measured daily.
- Context
- This is Coval's pooled corpus-level synthesized-speech WER for the exact Sonic 3.6 endpoint, 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. It measures output intelligibility, not STT accuracy. The tested voice is not identified by the public route, so this is not a voice-specific result. Results are from Coval's hosted Cartesia endpoint, not local or native performance.
Blind native-speaker preference share, English (US) vs Eleven v3 (Show measurement, test conditions, and source)
- Source value
- 92
- Score
- 92
- Metric
- Share of valid blind pairwise votes preferring Cartesia Sonic 3.6 for English (US) over Eleven v3
- Unit
- %
- Benchmark version
- Cartesia internal blind native-speaker preference report, article published 2026-08-27
- Category
- audio
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Cartesia Sonic 3.6 launch report
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Cartesia reports a blind pairwise preference test with native-speaker panels. For this exact locale row, Sonic 3.6 was compared with the named competitor using a comparable voice; the displayed share is the model's wins divided by total valid votes.
- Context
- Vendor-run preference study. Cartesia does not publish per-locale vote counts, confidence intervals, or an immutable serving revision; percentages should not be treated as a cross-locale ranking.
Blind native-speaker preference share, English (UK) vs Eleven v3 (Show measurement, test conditions, and source)
- Source value
- 72
- Score
- 72
- Metric
- Share of valid blind pairwise votes preferring Cartesia Sonic 3.6 for English (UK) over Eleven v3
- Unit
- %
- Benchmark version
- Cartesia internal blind native-speaker preference report, article published 2026-08-27
- Category
- audio
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Cartesia Sonic 3.6 launch report
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Cartesia reports a blind pairwise preference test with native-speaker panels. For this exact locale row, Sonic 3.6 was compared with the named competitor using a comparable voice; the displayed share is the model's wins divided by total valid votes.
- Context
- Vendor-run preference study. Cartesia does not publish per-locale vote counts, confidence intervals, or an immutable serving revision; percentages should not be treated as a cross-locale ranking.
Blind native-speaker preference share, English (Australia) vs Eleven v3 (Show measurement, test conditions, and source)
- Source value
- 77
- Score
- 77
- Metric
- Share of valid blind pairwise votes preferring Cartesia Sonic 3.6 for English (Australia) over Eleven v3
- Unit
- %
- Benchmark version
- Cartesia internal blind native-speaker preference report, article published 2026-08-27
- Category
- audio
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Cartesia Sonic 3.6 launch report
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Cartesia reports a blind pairwise preference test with native-speaker panels. For this exact locale row, Sonic 3.6 was compared with the named competitor using a comparable voice; the displayed share is the model's wins divided by total valid votes.
- Context
- Vendor-run preference study. Cartesia does not publish per-locale vote counts, confidence intervals, or an immutable serving revision; percentages should not be treated as a cross-locale ranking.
Blind native-speaker preference share, Spanish (Mexico) vs Eleven v3 (Show measurement, test conditions, and source)
- Source value
- 70
- Score
- 70
- Metric
- Share of valid blind pairwise votes preferring Cartesia Sonic 3.6 for Spanish (Mexico) over Eleven v3
- Unit
- %
- Benchmark version
- Cartesia internal blind native-speaker preference report, article published 2026-08-27
- Category
- audio
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Cartesia Sonic 3.6 launch report
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Cartesia reports a blind pairwise preference test with native-speaker panels. For this exact locale row, Sonic 3.6 was compared with the named competitor using a comparable voice; the displayed share is the model's wins divided by total valid votes.
- Context
- Vendor-run preference study. Cartesia does not publish per-locale vote counts, confidence intervals, or an immutable serving revision; percentages should not be treated as a cross-locale ranking.
Blind native-speaker preference share, Spanish (Spain) vs Eleven v3 (Show measurement, test conditions, and source)
- Source value
- 91
- Score
- 91
- Metric
- Share of valid blind pairwise votes preferring Cartesia Sonic 3.6 for Spanish (Spain) over Eleven v3
- Unit
- %
- Benchmark version
- Cartesia internal blind native-speaker preference report, article published 2026-08-27
- Category
- audio
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Cartesia Sonic 3.6 launch report
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Cartesia reports a blind pairwise preference test with native-speaker panels. For this exact locale row, Sonic 3.6 was compared with the named competitor using a comparable voice; the displayed share is the model's wins divided by total valid votes.
- Context
- Vendor-run preference study. Cartesia does not publish per-locale vote counts, confidence intervals, or an immutable serving revision; percentages should not be treated as a cross-locale ranking.
Blind native-speaker preference share, Portuguese (Brazil) vs Eleven v3 (Show measurement, test conditions, and source)
- Source value
- 67
- Score
- 67
- Metric
- Share of valid blind pairwise votes preferring Cartesia Sonic 3.6 for Portuguese (Brazil) over Eleven v3
- Unit
- %
- Benchmark version
- Cartesia internal blind native-speaker preference report, article published 2026-08-27
- Category
- audio
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Cartesia Sonic 3.6 launch report
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Cartesia reports a blind pairwise preference test with native-speaker panels. For this exact locale row, Sonic 3.6 was compared with the named competitor using a comparable voice; the displayed share is the model's wins divided by total valid votes.
- Context
- Vendor-run preference study. Cartesia does not publish per-locale vote counts, confidence intervals, or an immutable serving revision; percentages should not be treated as a cross-locale ranking.
Blind native-speaker preference share, French vs Eleven v3 (Show measurement, test conditions, and source)
- Source value
- 86
- Score
- 86
- Metric
- Share of valid blind pairwise votes preferring Cartesia Sonic 3.6 for French over Eleven v3
- Unit
- %
- Benchmark version
- Cartesia internal blind native-speaker preference report, article published 2026-08-27
- Category
- audio
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Cartesia Sonic 3.6 launch report
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Cartesia reports a blind pairwise preference test with native-speaker panels. For this exact locale row, Sonic 3.6 was compared with the named competitor using a comparable voice; the displayed share is the model's wins divided by total valid votes.
- Context
- Vendor-run preference study. Cartesia does not publish per-locale vote counts, confidence intervals, or an immutable serving revision; percentages should not be treated as a cross-locale ranking.
Blind native-speaker preference share, Hindi vs Gemini 3.1 (Show measurement, test conditions, and source)
- Source value
- 55
- Score
- 55
- Metric
- Share of valid blind pairwise votes preferring Cartesia Sonic 3.6 for Hindi over Gemini 3.1
- Unit
- %
- Benchmark version
- Cartesia internal blind native-speaker preference report, article published 2026-08-27
- Category
- audio
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Cartesia Sonic 3.6 launch report
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Cartesia reports a blind pairwise preference test with native-speaker panels. For this exact locale row, Sonic 3.6 was compared with the named competitor using a comparable voice; the displayed share is the model's wins divided by total valid votes.
- Context
- Vendor-run preference study. Cartesia does not publish per-locale vote counts, confidence intervals, or an immutable serving revision; percentages should not be treated as a cross-locale ranking.
Gradium hard-case TTS overall sentence pass rate (Show measurement, test conditions, and source)
- Source value
- 75.1
- Score
- 75.1
- Metric
- All-criteria share of sentences where every element is pronounced correctly and completely; higher is better
- Unit
- %
- Benchmark version
- Gradium hard-case TTS set, generated 2026-08; source article 2026-08-31; default settings
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Gradium, Gradium TTS: you no longer have to choose between latency and accuracy
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Gradium hard-case TTS evaluation generated in August 2026 with default settings. The 500-sentence set has 100 sentences per language in English, German, French, Spanish, and Portuguese. Independent native-speaker raters heard randomized, loudness-normalized outputs. The all-criteria sentence pass rate is pooled across ten criteria and averaged across the five languages with equal language weights. Voice IDs and immutable API revisions are not published. Exact table model: Cartesia Sonic 3.6.
- Context
- Gradium is a competing TTS vendor and reports its own comparison; source type is vendor-reported, not an independent leaderboard. The four comparison models were selected by Gradium. The score is macro-weighted across languages and criteria, so no single pooled sentence denominator or confidence interval is asserted. Voice IDs and serving revisions are unspecified.
Cartesia-reported maximum blind head-to-head listener preference vs Sonic-3.5 across 15 locales (Show measurement, test conditions, and source)
- Source value
- 93
- Score
- 93
- Metric
- Maximum reported share of blind head-to-head tests preferring Sonic-3.6 over Sonic-3.5; source says “up to” across 15 locales
- Unit
- %
- Category
- audio
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Cartesia TTS Research, Introducing Sonic-3.6
- Status
- active
- Retrieved at
- 2026-10-19
- Methodology
- Vendor-reported result. Cartesia's 2026-08-27 product-research post says listeners preferred Sonic-3.6 over Sonic-3.5 in up to 93% of blind head-to-head tests across fifteen locales. This is the publisher's maximum/ceiling claim, not a mean or overall 15-locale score.
- Context
- Vendor-reported, not an independent evaluation. The captured post does not identify which locale/test produced the “up to 93%” ceiling, preference-test N, confidence intervals, or full per-locale result table. The separate 92% US-English claim against Eleven v3 is already present in the baseline and is not added.
Cekura TTS hard-part accuracy: numbers and shorthand (Show measurement, test conditions, and source)
- Source value
- 71
- Score
- 71
- Metric
- Share of hard-part readings accepted for numbers and shorthand
- Unit
- %
- Benchmark version
- Cekura TTS Model Benchmarks; current page snapshot 2026-10-19; run date and serving revision not stated
- Category
- audio
- Direction
- Higher is better
- Source type
- official-leaderboard
- Evaluator
- Cekura Bench TTS Model Benchmarks
- Status
- active
- Retrieved at
- 2026-10-19
- Methodology
- Official benchmark leaderboard. Evaluator-run TTS benchmark. Cekura says each model read the same 104 agent sentences three times from one US East cloud region, using one stock voice and its own streaming protocol. Non-streaming timing runs from sending the full text to first audible sound; streaming timing runs from the last word sent to first audible sound. Reading accuracy is estimated from Deepgram Nova-3 and OpenAI Whisper transcripts, counting accepted readings without human review. Naturalness uses UTMOS, an open model predicting a listening-test score. The captured rows do not state per-cell N, CI, exact run date, serving/API revision, or exact voice ID unless shown separately.
- Context
- Cekura discloses that it runs the benchmark itself and sells testing and monitoring tools in the same market. The results are evaluator-published and include voice, runtime, protocol, ASR, and UTMOS effects; they are not a human-only or isolated-model evaluation.
Page 1 of 13
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 83 models
pyannote Community-1, pyannote: 73
Nemotron-3-Diarization, NVIDIA: 57
Gemini 3.5 Transcribe, Google: 52
Confucius4-R2T2, NetEase Youdao: 52
Granite Speech 5.0 470M TurboCTC, IBM: 52
Kokoro-82M, hexgrad: 51
Nova-3, Deepgram: 51
Fish Audio S2 Pro, Fish Audio: 51
NVIDIA Nemotron 3.5 Transcribe, NVIDIA: 50
Granite Speech 5.0 470M TurboCTC NC, IBM: 50
Orukeet, Oruk: 49
Lyria 3.5, Google: 47
Gemini 3.5 Live Translate, Google: 46
Phonon-2, Fermion Research: 44
Aura-2, Deepgram: 44
Gemini 3.1 Flash TTS, Google: 44
MiniMax-Music3, MiniMax: 41
Audio8-ASR-Infinite, Edge0 AI: 41
GPT-Realtime 2.1, OpenAI: 41
Sonic 3.6, Cartesia: 39
Sonic 3.6
As of November 2026 · Source: gradually.ai popularity index
Over time
Weekly values from October 18, 2026 to November 5, 2026
Sonic 3.6Eleven v3GPT-Live-TranscribeScribe v2Maya 2 Native
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
Sources and data date
Show all 43 sources
Benchmark sources at a glance
Which independent evaluators tested this model. This is coverage, not a quality ranking.
No active result from the reviewed external evaluators can currently be attributed to this model. Other source evidence remains available below.
Without vendor claims and historical or disputed results; those stay labelled in the single values.