Introducing VoicePing ASR Model V0.1, an Asian-language-focused speech-to-text model for Japanese, Korean, Chinese, Vietnamese, and English.
Introducing VoicePing ASR Model V0.1
Today we are introducing VoicePing ASR Model V0.1, our speech-to-text model focused on the Asian languages that matter most to VoicePing: Japanese, Korean, Chinese, and Vietnamese, alongside English.
VoicePing is built around spoken communication across Asia: meetings, events, voice translation, transcripts, summaries, and search. In those workflows, ASR is not an isolated feature. It is the first layer of the entire product experience. If the transcript is unstable, every downstream step becomes less useful.
VoicePing ASR Model V0.1 is designed for that reality. It focuses on Japanese, Korean, Chinese, Vietnamese, and English, with the goal of producing cleaner transcripts for real conversations.
One Model For Asian Languages And English
General-purpose speech recognition has improved quickly, but real speech across Asian languages and English still has difficult edges:
- Japanese, Korean, and Chinese need language-aware text handling.
- Vietnamese depends on accurate tone marks and word boundaries.
- Long or noisy clips can expose partial transcripts, empty outputs, and repeated text.
- Cloud models can behave differently across languages, even when the API looks uniform.
- A system that performs well on a public benchmark is not always the best fit for meetings, events, and voice translation.
VoicePing ASR Model V0.1 is our first consolidated model built around this Asian-language product surface. The benchmark below asks a practical question: how well does it transcribe the speech our users actually care about?
What It Does
VoicePing ASR Model V0.1 transcribes speech in:
- English
- Japanese
- Korean
- Chinese
- Vietnamese
The output is the transcript that powers later VoicePing features such as translation, captions, meeting notes, and searchable conversation history.
This article uses ASR and STT interchangeably. Both mean speech-to-text transcription.
Evaluations
Dataset
The evaluation uses an Asian-language-focused VoicePing speech set with 1,000 clips per language, about 41 hours of audio in total. The clips reflect the kind of speech VoicePing handles in practice: real conversations rather than clean read-aloud recordings.
| Language | Clips |
|---|---|
| English | 1,000 |
| Japanese | 1,000 |
| Vietnamese | 1,000 |
| Korean | 1,000 |
| Chinese | 1,000 |
| Total | 5,000 |
Every system is tested on the same audio set.
Models Compared
We compare VoicePing ASR Model V0.1 with widely used cloud speech systems and open ASR models, including Google Cloud STT, Azure AI Speech, OpenAI transcription models, ElevenLabs Scribe v2, Deepgram Nova-3, Qwen3-ASR, and SenseVoiceSmall.
Scoring
The headline metric is word error rate (WER): lower is better. WER measures how many words are inserted, deleted, or substituted compared with the human reference transcript.
Latency
Accuracy is not the only requirement for production ASR. We also measure how long each system takes to return a transcript, because a model that is accurate but slow can still feel poor in live meetings and events.
Main Results
The chart below compares the average word error rate across the five languages for VoicePing ASR Model V0.1 and the external speech-to-text systems in this benchmark. Lower bars are better. Per-language charts follow in the results section.
Accuracy by Language
| System | EN WER | JA WER | VI WER | KO WER | ZH WER | Macro WER |
|---|---|---|---|---|---|---|
| VoicePing ASR Model V0.1 | 20.2% | 20.4% | 15.5% | 24.5% | 16.0% | 19.3% |
| Google Cloud STT V1 default | 23.1% | 23.5% | 52.1% | 57.8% | 44.2% | 40.1% |
| Google Cloud STT Chirp 2 | 24.5% | 29.7% | 14.8% | 32.8% | 22.6% | 24.9% |
| Google Cloud STT Chirp 3 | 22.9% | 26.4% | 20.1% | 37.4% | 19.2% | 25.2% |
| Azure AI Speech | 23.0% | 21.1% | 21.0% | 37.3% | 22.5% | 25.0% |
| OpenAI GPT-4o Transcribe | 50.6% | 52.4% | 64.3% | 44.1% | 29.1% | 48.1% |
| OpenAI GPT Realtime Whisper | 31.8% | 26.2% | 20.9% | 33.0% | 22.4% | 26.9% |
| Qwen3-ASR 0.6B | 23.8% | 29.7% | 26.2% | 38.2% | 20.9% | 27.7% |
| Qwen3-ASR 1.7B | 21.9% | 25.0% | 22.0% | 33.1% | 20.0% | 24.4% |
| SenseVoiceSmall | 28.0% | 37.4% | 99.9% | 45.9% | 28.1% | 47.9% |
| ElevenLabs Scribe v2 | 28.6% | 20.3% | 15.4% | 31.5% | 21.2% | 23.4% |
| Deepgram Nova-3 | 29.3% | 28.0% | 38.4% | 44.8% | 29.2% | 34.0% |
Leaderboard
| System | Macro WER | Median latency | Notes |
|---|---|---|---|
| VoicePing ASR Model V0.1 | 19.3% | 1.22s | VoicePing Asian-language ASR |
| Google Cloud STT V1 default | 40.1% | 7.47s | Cloud speech-to-text |
| Google Cloud STT Chirp 2 | 24.9% | 7.12s | Cloud speech-to-text |
| Google Cloud STT Chirp 3 | 25.2% | 7.32s | Cloud speech-to-text |
| Azure AI Speech | 25.0% | 7.12s | Cloud speech-to-text |
| OpenAI GPT-4o Transcribe | 48.1% | 1.53s | OpenAI transcription |
| OpenAI GPT Realtime Whisper | 26.9% | 7.17s | OpenAI transcription |
| Qwen3-ASR 0.6B | 27.7% | 3.56s | Open ASR model |
| Qwen3-ASR 1.7B | 24.4% | 4.18s | Open ASR model |
| SenseVoiceSmall | 47.9% | 0.07s | Open ASR model |
| ElevenLabs Scribe v2 | 23.4% | 3.07s | Cloud speech-to-text |
| Deepgram Nova-3 | 34.0% | 1.33s | Cloud speech-to-text |
VoicePing ASR Model V0.1 combines the lowest macro WER in this benchmark with one of the fastest median response times, at 1.22 seconds. The systems that respond faster in this run give up a large amount of accuracy to do so.
Results by Language
English
VoicePing ASR Model V0.1 has the lowest English WER in this comparison at 20.2%, ahead of Qwen3-ASR 1.7B and the major cloud systems tested here.
Japanese
Japanese is one of the most important languages for VoicePing. On this dataset, VoicePing ASR Model V0.1 reaches 20.4% WER, in a virtual tie with ElevenLabs Scribe v2 (20.3%) for the best Japanese result, ahead of Azure AI Speech and the open ASR baselines.
Vietnamese
Vietnamese is the closest race in this benchmark: Google Cloud STT Chirp 2 leads at 14.8% WER, with ElevenLabs Scribe v2 (15.4%) and VoicePing ASR Model V0.1 (15.5%) essentially tied just behind.
Korean
Korean shows one of the clearest gaps in the benchmark. VoicePing ASR Model V0.1 records 24.5% WER, well ahead of the next group of systems.
Chinese
Chinese is another strong area for VoicePing ASR Model V0.1, at 16.0% WER, with Google Cloud STT Chirp 3 and Qwen3-ASR 1.7B closest behind.
What We Learned
- VoicePing ASR Model V0.1 has the strongest overall accuracy in this benchmark, with 19.3% macro WER across the five languages.
- The result is not uniform by language: English, Korean, and Chinese show the clearest VoicePing advantages in this run, Japanese is a virtual tie with the best cloud system, and Vietnamese is a close race decided by less than one point.
- Larger general-purpose systems do not automatically win on product-specific Asian-language speech.
- Accuracy and response time both matter for the user experience, especially in live meetings and events.
What’s Next
VoicePing ASR Model V0.1 is a first release, and this benchmark is a snapshot. The dataset is built from the kind of speech VoicePing handles in practice, so it measures readiness for our product rather than standing in for a universal public benchmark — and the cloud systems in the comparison will keep evolving, as will our model. Latency also depends on the deployment environment, so treat the speed numbers as indicative rather than absolute.
From here, our work focuses on the places this evaluation points to: reducing the error patterns that remain in each language, extending the test set with noisier, longer, and more domain-specific audio, and broadening the comparison as new speech-to-text systems ship. Automatic scores guide that work, and human transcript review stays part of every release decision.
Conclusion
VoicePing ASR Model V0.1 is our first consolidated ASR model focused on Japanese, Korean, Chinese, and Vietnamese, while also supporting English for real product workflows. In this 5,000-clip benchmark it delivers the strongest overall accuracy of the systems tested, at some of the fastest response times — and it is the transcription layer every other VoicePing feature builds on.
The important shift is focus: we are evaluating ASR as part of a real communication product for Asian languages and English, not as an isolated model demo.
