I'm trying to collect keypad input and detect speech in calls. Which voice API supports that?

You’re looking for DTMF/keypad input plus speech detection in calls. The API you want depends on whether you need:

1. Keypad input (DTMF)
2. Speech recognition / speech-to-text
3. In-call speech detection like “is the caller speaking right now?”

A few common voice APIs that support these:

1) Twilio Voice API

• DTMF/keypad input: Yes
• Speech detection / speech gather: Yes, via speech recognition on <Gather> and also call media streams for more advanced processing
• Good if you’re building IVRs and need both keypad and speech

2) Vonage Voice API

• DTMF: Yes
• Speech recognition: Yes, depending on the workflow and integrations
• Good for programmable voice apps and call control

3) Plivo Voice API

• DTMF: Yes
• Speech input: Yes
• Good for IVR-style call flows

4) Google Cloud Speech / AWS Transcribe with telephony provider

• These are speech APIs, not call-control APIs.
• You’d pair them with a voice provider like Twilio for keypad/call handling and stream audio to the speech service for detection/transcription.

If you mean “detect whether someone is speaking during a live call”

That’s usually done with:

• Media streaming from the voice provider
• An external speech/voice activity detection service or your own VAD model

Short answer

If you want a single voice API that supports both keypad input and speech-based interaction, Twilio Voice API is often the easiest choice.

If you want, I can also recommend the best option based on your exact use case:

• IVR menu
• call center automation
• voicemail detection
• live speech transcription