Product Overview
The Tongyi Tingwu Intelligent Minutes Agent uses speech recognition and large language models to summarize and analyze real-time or offline recordings from scenarios such as meetings, interviews, and classes.
Benefits
High-accuracy speech recognition- Transcribes speech from real-time audio streams or audio/video files into text. This feature supports transcription and real-time bidirectional translation for Chinese, English, Cantonese, Japanese, Korean, German, French, and Russian.
- Transcription results include paragraph and sentence divisions, and word-level start and end times for displaying corresponding captions.
- Supports features such as chapter overviews, large model summaries (full-text summary, speaker summary, key point review, mind map), key point extraction (keywords, to-do items, scenario detection), PPT extraction and summary, spoken-to-written language conversion, and custom prompts.
Scenarios
Designed for business scenarios such as smart offices, education and training, finance, and media.
Billing rules
Audio and video (billed by duration)
Model | Feature | Is Optional | Standard Price |
|---|---|---|---|
ASR |
| Select one of the two | CNY 0.6/hour |
Large model |
| Each feature is optional and billed independently based on the duration of the processed audio or video. If you use multiple features or prompts, the charges are cumulative. For example, using both keywords and chapter overview costs CNY 0.064 + CNY 0.064 = CNY 0.128/hour. | CNY 0.064/hour |
Multimodal |
| Optional | CNY 0.64/hour |
Translation |
| Optional | CNY 4/hour CNY 0.5/hour |
Plain text processing
CNY 0.95/1,000 calls
NoteBilling is based on the input text volume. Every 150 tokens of input are counted as one call. An input with fewer than 150 tokens is also counted as one call. For example, an input of 400 tokens is counted as 3 calls.