湘语体系
90.1
U2-ASR
86.2
FunASR
1.5
83.4
Qwen3-
ASR-1.7B
80.4
Seed-
ASR
U2-ASR
Beyond hearing, truly understanding expression
100+ dialects to many languages · one-shot · real-time · long audio
U2-ASR
Built on full-scenario capabilities, U2-ASR combines contextual understanding and custom vocabulary optimization to move beyond rigid word-by-word recognition into more natural semantic understanding. It covers 100+ dialects and multiple languages while balancing accuracy with clean, normalized text output. One-shot recognition and real-time transcription support instant, low-latency interactions—ready for personal, study, office, and everyday use.
90%
Accuracy in Complex Noise
99.2%
AISHELL-1 Test Set Score
98.4%
AISHELL-3 Test Set Score
98.4%
Libri Clean Test Set Score
Excels on industrial-grade dialect benchmarks, outperforming mainstream ASR overall.
U2-ASROther Models
客家话体系
83.4
U2-ASR
65.0
FunASR
1.5
42.6
Qwen3-
ASR-1.7B
71.8
Seed-
ASR
闽语体系
84.9
U2-ASR
68.2
FunASR
1.5
73.1
Qwen3-
ASR-1.7B
71.5
Seed-
ASR
赣语方言
88.3
U2-ASR
64.6
FunASR
1.5
64.0
Qwen3-
ASR-1.7B
74.1
Seed-
ASR
晋语体系
89.6
U2-ASR
73.1
FunASR
1.5
73.0
Qwen3-
ASR-1.7B
77.4
Seed-
ASR
吴语体系（苏州话）
82.2
U2-ASR
62.0
FunASR
1.5
53.0
Qwen3-
ASR-1.7B
50.7
Seed-
ASR
西南官话（武汉话）
92.1
U2-ASR
89.2
FunASR
1.5
85.8
Qwen3-
ASR-1.7B
85.6
Seed-
ASR
冀鲁官话
96.2
U2-ASR
93.0
FunASR
1.5
93.7
Qwen3-
ASR-1.7B
93.0
Seed-
ASR
西南官话（四川话）
94.7
U2-ASR
89.2
FunASR
1.5
90.2
Qwen3-
ASR-1.7B
90.1
Seed-
ASR
吴语体系（上海话）
89.0
U2-ASR
74.6
FunASR
1.5
72.5
Qwen3-
ASR-1.7B
79.8
Seed-
ASR
粤语体系
93.0
U2-ASR
90.2
FunASR
1.5
89.9
Qwen3-
ASR-1.7B
87.7
Seed-
ASR
Core Capabilities
Core Capabilities
Cross-dialect bilingual interoperability that understands local accents
Chinese, English, and 15 int'l languages (Arabic, German, Spanish, French, Indonesian, Japanese, Korean, Portuguese, Russian, Turkish, Vietnamese, Thai, Italian), plus 100+ dialects incl. Cantonese, Sichuanese, Shanghainese, Hokkien—full seven major Chinese dialect coverage. Accurate for single dialect, multilingual mix, and cross-border use.
One-shot fast recognition—instant speech-to-text in seconds
Low-latency, high-efficiency response—single utterances convert to text instantly, ideal for voice commands, quick Q&A, and other real-time interaction scenarios.
Context-aware understanding that goes beyond transcription
Moves beyond mechanical word-by-word transcription, using conversation context and specialized vocabularies for deeper semantic understanding. Handles noisy environments, strong accents, and professional jargon—faithfully restoring what was meant.
Full-scenario transcription + structured output for short and long audio
Supports real-time streaming transcription and batch processing of long audio—up to 5 hours—for live note-taking and efficient archiving. Includes speaker diarization, timestamps, and smart punctuation/sentence splitting for clean, ready-to-use text with no extra editing.
Highlights
HighlightsStrong Noise Robustness
Optimized for real recording environments, maintaining stable recognition under complex background sounds, noisy malls, and meeting rooms.
Broad Dialect and Language Coverage
Beyond Mandarin: 100+ dialects and 15 languages across Asia, Europe, Middle East, and Latin America. One model for domestic cross-region and overseas transcription—no per-language procurement.
Stronger Domain Semantic Understanding
Context and hotword injection improve recognition of domain terms in medical, automotive, and customer-service workflows.
Accurate Speaker-Separated Transcription
Speaker diarization, smart segmentation, punctuation prediction, and timestamps make transcripts clearer and easier to organize.
Use Cases
Use CasesOffice Document Input
Generate work docs, emails, and draft plans quickly from speech to speed up content input.
Medical Record Entry
Recognize large volumes of medical terminology in real time and support physician dictation for faster EMR creation.
Communication and Translation
Accurately restores and clearly records different accents and ways of expression.
Meeting Audio Transcription
Parse long recordings in one click and output structured, well-formatted transcripts with less manual effort.
Capabilities
CapabilitiesAuto language detection: transcribes without a specified language—handles real-world scenarios with unknown input language.
15 int'l languages: Asia, Europe, Middle East, and Latin America for cross-border multilingual transcription.
Supports audio-to-text for long-form scenarios such as meetings, lectures, customer service, and business recordings.
Supports speaker diarization to distinguish segments from different speakers.
Supports smart sentence splitting and punctuation prediction to improve readability.
Supports timestamp output for subtitles, search, and audio-video alignment.
Supports context and hotword enhancement to improve recognition of proper nouns and industry terms.
Supports one-shot recognition with millisecond response for instant interaction.
Supports real-time streaming transcription with low-latency sync as you speak.
Flexible pricing, custom solutions, private deployment
U2-ASR: Beyond hearing, truly understanding expression
Built on full-scenario capabilities, U2-ASR combines contextual understanding and custom vocabulary optimization to move beyond rigid word-by-word recognition into more natural semantic understanding. It covers 100+ dialects and multiple languages while balancing accuracy with clean, normalized text output. One-shot recognition and real-time transcription support instant, low-latency interactions—ready for personal, study, office, and everyday use.
90%
Accuracy in Complex Noise
99.2%
AISHELL-1 Test Set Score
98.4%
AISHELL-3 Test Set Score
98.4%
Libri Clean Test Set Score
Excels on industrial-grade dialect benchmarks, outperforming mainstream ASR overall.
U2-ASROther Models
Xiang
90.1
U2-ASR
86.2
FunASR
1.5
83.4
Qwen3-
ASR-1.7B
80.4
Seed-
ASR
Hakka
83.4
U2-ASR
65.0
FunASR
1.5
42.6
Qwen3-
ASR-1.7B
71.8
Seed-
ASR
Min
84.9
U2-ASR
68.2
FunASR
1.5
73.1
Qwen3-
ASR-1.7B
71.5
Seed-
ASR
Gan
88.3
U2-ASR
64.6
FunASR
1.5
64.0
Qwen3-
ASR-1.7B
74.1
Seed-
ASR
Jin
89.6
U2-ASR
73.1
FunASR
1.5
73.0
Qwen3-
ASR-1.7B
77.4
Seed-
ASR
Wu (Suzhou)
82.2
U2-ASR
62.0
FunASR
1.5
53.0
Qwen3-
ASR-1.7B
50.7
Seed-
ASR
SW (Wuhan)
92.1
U2-ASR
89.2
FunASR
1.5
85.8
Qwen3-
ASR-1.7B
85.6
Seed-
ASR
Ji-Lu
96.2
U2-ASR
93.0
FunASR
1.5
93.7
Qwen3-
ASR-1.7B
93.0
Seed-
ASR
SW (Sichuan)
94.7
U2-ASR
89.2
FunASR
1.5
90.2
Qwen3-
ASR-1.7B
90.1
Seed-
ASR
Wu (Shanghai)
89.0
U2-ASR
74.6
FunASR
1.5
72.5
Qwen3-
ASR-1.7B
79.8
Seed-
ASR
Yue
93.0
U2-ASR
90.2
FunASR
1.5
89.9
Qwen3-
ASR-1.7B
87.7
Seed-
ASR
Core Capabilities
Cross-dialect bilingual interoperability that understands local accents
Chinese, English, and 15 int'l languages (Arabic, German, Spanish, French, Indonesian, Japanese, Korean, Portuguese, Russian, Turkish, Vietnamese, Thai, Italian), plus 100+ dialects incl. Cantonese, Sichuanese, Shanghainese, Hokkien—full seven major Chinese dialect coverage. Accurate for single dialect, multilingual mix, and cross-border use.
One-shot fast recognition—instant speech-to-text in seconds
Low-latency, high-efficiency response—single utterances convert to text instantly, ideal for voice commands, quick Q&A, and other real-time interaction scenarios.
Context-aware understanding that goes beyond transcription
Moves beyond mechanical word-by-word transcription, using conversation context and specialized vocabularies for deeper semantic understanding. Handles noisy environments, strong accents, and professional jargon—faithfully restoring what was meant.
Full-scenario transcription + structured output for short and long audio
Supports real-time streaming transcription and batch processing of long audio—up to 5 hours—for live note-taking and efficient archiving. Includes speaker diarization, timestamps, and smart punctuation/sentence splitting for clean, ready-to-use text with no extra editing.
Highlights
Strong Noise Robustness
Optimized for real recording environments, maintaining stable recognition under complex background sounds, noisy malls, and meeting rooms.
Broad Dialect and Language Coverage
Beyond Mandarin: 100+ dialects and 15 languages across Asia, Europe, Middle East, and Latin America. One model for domestic cross-region and overseas transcription—no per-language procurement.
Stronger Domain Semantic Understanding
Context and hotword injection improve recognition of domain terms in medical, automotive, and customer-service workflows.
Accurate Speaker-Separated Transcription
Speaker diarization, smart segmentation, punctuation prediction, and timestamps make transcripts clearer and easier to organize.
Use Cases
Office Document Input
Generate work docs, emails, and draft plans quickly from speech to speed up content input.
Medical Record Entry
Recognize large volumes of medical terminology in real time and support physician dictation for faster EMR creation.
Communication and Translation
Accurately restores and clearly records different accents and ways of expression.
Meeting Audio Transcription
Parse long recordings in one click and output structured, well-formatted transcripts with less manual effort.
Capabilities
- Auto language detection: transcribes without a specified language—handles real-world scenarios with unknown input language.
- 15 int'l languages: Asia, Europe, Middle East, and Latin America for cross-border multilingual transcription.
- Supports audio-to-text for long-form scenarios such as meetings, lectures, customer service, and business recordings.
- Supports speaker diarization to distinguish segments from different speakers.
- Supports smart sentence splitting and punctuation prediction to improve readability.
- Supports timestamp output for subtitles, search, and audio-video alignment.
- Supports context and hotword enhancement to improve recognition of proper nouns and industry terms.
- Supports one-shot recognition with millisecond response for instant interaction.
- Supports real-time streaming transcription with low-latency sync as you speak.
