Overview
Fun-ASR-Flash is a non-real-time automatic speech recognition model that transcribes audio into text. Alibaba Cloud lists this snapshot under June 17, 2026; the snapshot ID contains the date June 15, 2026. (alibabacloud.com)
Capabilities and uses
- Supports 30 languages, including Chinese, English, Japanese, Korean, Vietnamese, Arabic, French and Spanish. It covers major Chinese dialect groups and more than 20 regional Mandarin accents.
- Supports contextual prompts and is optimized for classical Chinese poetry, punctuation prediction and normalization of numbers, dates and amounts.
- Suitable for short-recording transcription, cultural heritage, education and audiobooks. (docs.qwencloud.com)
Limits
- Audio files are limited to 5 minutes and 2 GB. This is an offline recognition model, not the separate real-time model.
- Function calling, structured output, web search, context caching, batch inference and fine-tuning are listed as unsupported. The official model page does not specify a token context-window limit. (help.aliyun.com)
Pricing
Checked November 16, 2026. Prices are billed by input audio duration; text output is free. These are separate official regional and platform listings, not currency conversions.
- Alibaba Cloud China site, CNY: Beijing: CNY 0.00022 per input audio second; Singapore international deployment: CNY 0.00026 per input audio second. Beijing lists 36,000 seconds of free quota, valid for 90 days from the latest of Model Studio activation, model release or application approval; the page lists no free quota for other regions. Official China pricing (help.aliyun.com)
- Alibaba Cloud international site, USD: Singapore international deployment: USD 0.000035 per input audio second; Beijing: USD 0.00003 per input audio second. Singapore lists 36,000 seconds of free quota under the same 90-day timing rule; the page lists no free quota for other regions. Official international pricing (alibabacloud.com)