Overview
A non-real-time automatic speech recognition model for transcribing complete audio and video files into text. It is intended for uses such as video subtitles, customer-service quality inspection, and financial recording review. (docs.qwencloud.com)
Capabilities
- Supports multilingual transcription, Chinese dialects and regional accents. The release announcement lists 30 languages, including Chinese, English, Japanese and Korean.
- Supports hotwords and contextual input to improve recognition of specialized terms, as well as punctuation prediction and text normalization.
- Supports speaker diarization; emotion recognition is not supported. (alibabacloud.com)
Input and limitations
- Offline file transcription through HTTP; one publicly accessible file URL per request.
- Maximum file size is 2 GB and maximum duration is 12 hours. When speaker diarization is enabled, the documentation recommends a maximum of 2 hours.
- Supported formats include AAC, AMR, FLAC, M4A, MP3, MP4, OGG, OPUS, WAV and WEBM; the documented sample rate is unrestricted.
- This is not the real-time streaming variant. (help.aliyun.com)
Pricing
Official listed prices were checked on November 16, 2026. Billing is based on input audio duration; text output is free. These are separate regional and platform listings, not currency conversions.
| Official listing | Region and scope | Input audio | Output |
|---|---|---|---|
| Mainland China Model Studio | China North 2 (Beijing) | CNY 0.00022 per second | Free |
| Mainland China Model Studio | Singapore, international deployment | CNY 0.00026 per second | Free |
| International Model Studio | Singapore, international deployment | USD 0.000035 per second | Free |
| International Model Studio | China North 2 (Beijing) | USD 0.000032 per second | Free |
The current price tables list no free quota for this model in the Beijing mainland listing or Singapore international listing. Mainland China official pricing; international official pricing. (help.aliyun.com)
Release and official sources
- Release date: July 30, 2026, as listed for the Singapore international deployment in the official model release announcement. (alibabacloud.com)
- Official speech recognition model documentation and official model description.