The request names AssemblyAI and Activepieces, but none of the candidate products is AssemblyAI or Activepieces, so the answer cannot be about them directly; the closest matches are speech-to-text APIs that could be used for transcription in an Activepieces workflow. Deepgram 5 and ElevenLabs 2 offer transcription APIs with documented actions, and Rev AI 3 provides an asynchronous transcription API. Whether you want a replacement for AssemblyAI or an Activepieces-compatible integration is
A question back: Are you looking for an alternative to AssemblyAI, or a speech-to-text step that works inside an Activepieces flow?
- 1
ElevenLabselevenlabs.ioverified and signed
Transcribes audio to text with Scribe v2 via its speech-to-text action and documented API 2.
official:MCP server · ElevenLabsAPI · ElevenLabs API Documentation+59 community tools built for it
for agents: speech-to-text · speech_to_text · Text_to_speech_v1_text_to_speech__voice_id__post
- 2
Deepgramdeepgram.comverified and signed
Transcribes audio or video files and URLs into transcripts with optional speakers and paragraphs, and offers a callback-based transcription action 5.
+6 community tools built for it
for agents: transcribe_audio · create_transcription_callback · list_models
- 3
Rev AI Asynchronous Speech-To-Text APIrev.aiverified and signed
Submits pre-recorded files for transcription through a job submission action and retrieves transcripts and captions 3.
official:API · Asynchronous Speech-To-Text API Documentation
for agents: GetTranscriptById · SubmitTranscriptionJob · GetCaptions
Sources
- 2Free AI Voice Generator & Voice Agents Platform | ElevenLabselevenlabs.io · read 2026-10-29
- 3The Most Accurate Speech-To-Text API | Rev AIrev.ai · read 2026-10-29
- 5Enterprise voice AI: STT, TTS & agent APIs | Deepgramdeepgram.com · read 2026-10-29
For agents: the 10 actions behind this answer
transcribe_url · @speechweave/mcp found in code, not yet checked
Start transcription from a public HTTPS URL and wait until it finishes. Prefer for short/medium media when you need the transcript in this turn. Set task='translate' to get an English translation instead of a same-language transcript. On wait timeout, returns job_id, then call get_job_status. For lo
speech-to-text · speech-to-text verified and signed
Transcribe audio to text using ElevenLabs Scribe v2. Use when converting audio/video to text, generating subtitles, transcribing meetings, or processing spoken content.
GetTranscriptById · Asynchronous Speech-To-Text API Documentation verified and signed
Get Transcript By Id
Transcribe · Azure Speech Transcription API verified and signed
Transcribes the provided audio stream.
transcribe_audio · Deepgram (Activepieces) verified and signed
Transcribe Audio — Transcribes an audio or video file or URL and returns the transcript, with optional speakers, paragraphs, summary, topics and more.
transcribe_audio · LocalAI (Activepieces) verified and signed
Transcribe Audio — Turn speech in an audio file into text.
· Uttera website read, tools not yet checked
Transcribe, summarize, translate, speak, analyse pronunciation, and generate music and sound effects. OpenAI-compatible. Sustainable AI, processed entirely in Spain, on solar power.
vocametrix_transcribe_audio · vocametrix found in code, not yet checked
Transcribe an audio file using Azure Speech-to-Text with streaming progress.
Speech to Text (Whisper Transcription) · agentsvc.io verified and signed
Speech to text for agents: transcribe a voice note, meeting, podcast or call recording into text with timestamps, in about 100 languages with automatic language detection. Runs OpenAI Whisper (small by default, base for speed) on agentsvc.io itself, the audio is deleted after the call. Send audio_ur
Transcribe a recording · Scriptivox verified and signed
Turn a recorded audio or video file into text, with optional speaker diarization and word-level timestamps. Not for live or real-time audio. e.g. Transcribe this interview and label who is speaking. e.g. Generate SRT captions for this video.
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.