
Descript is an AI-powered video and audio editor that lets you cut recordings by editing their transcript, then clean up, caption, translate and repurpose the result. Underlord, its AI co-editor, plus Studio Sound, voice cloning and screen recording turn a single tool into a full production pipeline for creators and teams.
| 维度 / 指标 | ![]() | |
|---|---|---|
| 官方域名 | deepseek.com | descript.com |
| 定价模式 | 免费增值 | 免费增值,付费 |
| 月访问量 | 3.73亿 (372,761,591) | 326.78万 (3,267,817) |
| 月度环比流量变化 | +1536.22万 (+4.3%) | +32.69万 (+11.1%) |
| 主要市场与份额 | 🇨🇳 中国 (50.5%) | 🇺🇸 美国 (38.9%) |
| 直接访问 / 月 | 2.92亿 | 164.94万 |
| 用户评分 | 4.8 | 4.6 |
| 支持的平台 | 网站 | 网站 |
由 DeepSeek-V4 与 R1 驱动的高效开源权重与托管推理 AI,token 成本极低。
依托稀疏混合专家架构(V4-Pro 为 1.6T 中的 49B 激活参数),以极低延迟与内存开销处理最多 100 万 token。
由 DeepSeek-R1 开创并融入 V3.2 与 V4 思考模式,对复杂数学、逻辑与代码调试展示逐步自校验过程。
缓存命中时,输入 token 成本降至 Flash 的每百万 0.003 美元与 V4-Pro 的每百万 0.022 美元,非高峰 UTC 时段费率再低 50%。
在 chat.deepseek.com 及 iOS/Android 上免费提供文档上传、推理模式与网页搜索,核心模型不必付费墙。

Descript is an AI-powered video and audio editor that lets you cut recordings by editing their transcript, then clean up, caption, translate and repurpose the result. Underlord, its AI co-editor, plus Studio Sound, voice cloning and screen recording turn a single tool into a full production pipeline for creators and teams.
Edit audio and video by deleting, rearranging or retyping words in the automatic transcript, with a full multitrack timeline available when you need it.
Describe the edit you want in plain language and let the AI assistant assemble a tight cut, add clips and handle repetitive production work.
Remove background noise, level out voices, and fix misread lines with Regenerate Speech using a clone of your own voice.
Generate subtitles and transcripts in a click, translate video into other languages, and reformat long recordings into shareable vertical clips.

提供巨大的上下文窗口与深度的 Google 生态系整合,但对于纯推理任务感觉较不够专精。
提供巨大的上下文窗口与深度的 Google 生态系整合,但对于纯推理任务感觉较不够专精。
专项更强Provides industry-leading voice cloning and text-to-speech generation, but does not offer media recording or timeline editing.
专项更强Provides industry-leading voice cloning and text-to-speech generation, but does not offer media recording or timeline editing.

免费 / 开源提供同等级的开源权重基础模型与多语言效能,但需要自行架构主机或第三方推理设定。
免费 / 开源Enables running open-source models locally on consumer hardware, but lacks DeepSeek's cloud-scale hosted reasoning power.
免费 / 开源Enables running open-source models locally on consumer hardware, but lacks DeepSeek's cloud-scale hosted reasoning power.