Deepgram
来自 Deepgram · deepgram.com ↗
语音识别与语音 AI 的 API，可用于实时转写与音频理解。
speech-to-text api voice-ai
- 分类
- 音频、语音与音乐
- 商业模式
- 付费
- 可用性
- 全球
- 发布
- 2015
- 记录更新
- 2026-07-29
- 成立时间
- 2015
- 总部所在地
- US
概述
Deepgram is a speech recognition and voice AI API company focused on real-time transcription and audio understanding. It offers low-latency streaming speech-to-text, text-to-speech, and audio-intelligence models aimed at voice agents and contact centers. Pricing is usage-based, with free starter credits.
主要功能
- Low-latency streaming speech-to-text
- Text-to-speech and voice-agent APIs
- Audio intelligence such as summarization
- Self-hosted deployment options for enterprises
适用场景
- Real-time transcription for live applications
- Voice agents and IVR systems
- Contact-center analytics
定价
Pricing is usage-based per minute of audio; new developers receive free credits, with enterprise and self-hosted options.
常见问题
What does Deepgram do?
Deepgram provides speech recognition, text-to-speech, and voice-agent APIs, with a focus on real-time use.
Is Deepgram good for real-time use?
Yes. Deepgram emphasizes low-latency streaming transcription for live applications and voice agents.
How is Deepgram priced?
Deepgram is usage-priced per minute of audio, with free starter credits.
同类产品
- AIVA — AI 音乐创作助手，可为媒体项目生成多种风格的原创曲目。
- AssemblyAI — 语音转文本 API，为开发者提供转写与音频智能分析模型。
- Cartesia — 基于状态空间模型的实时语音生成平台，可实现低延迟语音合成。
- Descript — 由 AI 驱动的音视频编辑器，可通过编辑文字稿的方式来剪辑录音与视频。
- ElevenLabs — 提供文本转语音、声音克隆与配音的 AI 语音平台。
- Murf AI — AI 文本转语音平台，提供录音棚级合成语音，用于配音与旁白。
来源
本记录最近复核于 2026-07-29。
机器可读记录：/api/products/deepgram.json · 发现错误？提交纠错