# Record a call, both sides, and fetch the file or transcript afterwards

> Record a call, both sides, and fetch the file or transcript afterwards: 0 companies fully do it, 12 with a limit, 28 carry the tag. Every listing in Phone calling & voice transport with the reason and a source.

- Canonical: https://www.anchorterminal.com/niches/voice-calling.record-and-retrieve/
- Markdown: https://www.anchorterminal.com/niches/voice-calling.record-and-retrieve/index.md (~1,600 tokens)
- Slim: https://www.anchorterminal.com/niches/voice-calling.record-and-retrieve/index.min.md (~380 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/niches/voice-calling.record-and-retrieve/index.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-15

No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.

| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |
| --- | --- | --- | --- | --- |
| 0 | 12 | 28 | 11 | 6 |

## What counts as done

Recording can be switched on for a call that also carries a bidirectional agent stream, captures both parties (dual-channel available), excludes hold music when asked, sends a signed recording-ready callback, and the file is downloadable by id through the API with its retention period and any per-minute recording or storage price stated in docs.

How agents ask for it: record every call; download the call audio; get the transcript of yesterday's calls; save both sides of the conversation

## Does the job, with a limit

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

- **46elks Voice API (46elks AB)**: recordcall records the entire call on any action and posts a WAV link, fetchable for 72 hours only. No dual-channel option, no signed callback (IP allow list), and nothing says it works on websocket numbers.
- **Bandwidth Voice API + MCP (Bandwidth Inc.)**: StartRecording runs beside a stream with multiChannel for two channels and a Recording Available webhook. Callbacks carry Basic auth, not a signature, and no retention period was found; hold-music exclusion not documented.
- **didlogic Voice API (DID Logic PTE. LTD.)**: GET /api/v1/calls/recording/{callid} redirects to a signed S3 link and click2call carries recording_url, but recording must be enabled on the account, lookups are capped at 1 a minute, retention is per account and unstated, and no dual-channel option was found.
- **Exotel Voice API + MCP (Exotel Techcom Private Limited)**: The Voicebot applet has a Record option and the v3 call API takes channels dual; RecordingUrl arrives in StatusCallback. No signed callback, retention period or recording price in the docs read.
- **Infobip Calls API + MCP (Infobip Ltd)**: Calls, dialogs and conferences can be recorded with individual per-participant files or composed, downloaded by file id as WAV or pushed by SFTP. No retention period found in the Calls docs read.
- **jambonz (FirstFive8, Inc.)**: Recording starts by config verb or update call and goes by SIPREC or to your own S3, Azure or Google bucket; jambonz keeps no file to fetch by id and states no retention, and no ready callback was found.
- **Plivo Voice API (Plivo Inc.)**: Record API and Record XML capture calls up to 24 hours and the Recordings API lists and fetches files; the default recording cap is 60 seconds unless the API is used. Dual-channel with a live stream and hold-music exclusion were not confirmed.
- **SignalWire Voice API (SignalWire, Inc.)**: Recording runs in the background by SWML, Relay or calling.record with direction both and stereo, with status callbacks. Retention period not found in the guide.
- **Sinch Voice API + MCP (Sinch AB (publ))**: startRecording takes COMBINED, INBOUND or OUTBOUND; no separate-channel option, and no retention period found in the pages read.
- **Twilio Programmable Voice API + MCP (Twilio Inc.)**: The Recordings resource starts, pauses and deletes recordings on an in-progress call, stores them dual-channel by default and posts status callbacks. acc-08 found no documented way with a bidirectional stream, acc-18 captured hold music, and no default retention period was found.
- **Vonage Voice API + MCP (Vonage Holdings Corp.)**: record and conversation record take split conversation and up to 32 channels, with an eventUrl callback carrying the recording, and webhooks can be signed with an HS256 JWT. Retention period and hold-music exclusion not found.
- **Voximplant (Voximplant, Inc.)**: Call recording has a stereo option and an expire setting of 3 or 6 months or 1 to 3 years, with the link in RecordStarted; secure recordings need an authorised fetch. Hold-music exclusion and a signed ready callback were not found.

## Couldn't check

The docs were closed to us, or silent. These carry no mark either way.

- **LiveKit Telephony (LiveKit Incorporated)**: Recording runs through LiveKit Egress, outside the telephony pages read; SIPREC is marked not supported.
- **Telnyx Voice API + MCP (Telnyx LLC)**: Transfer accepts record-from-answer, but the recording commands, channels and retention pages were not read.

- not this job: DIDWW Voice and Number Management API (DIDWW Ireland Limited): No recording resource in API v3; only CDR exports.

## What stops agents finishing

- recording not available alongside a bidirectional stream
- a BAA, and with it recording of health calls, only on an enterprise plan (acc-41)
- recording files deleted after an undocumented period
- call-recording consent rules by country left to the caller

## Around it

Before: voice-calling.number-to-agent-stream, voice-calling.place-outbound-call. After: speech-to-text.long-recording-batch, speech-to-text.meeting-with-speakers, crm.log-activity.

## Where this comes from

Weight 11 of 100 in Phone calling & voice transport: 6 of 49 accounts: recording a bidirectional stream, recording both sides of an IVR bot, transcribing a dialled call, hold music in conference recordings, dual-channel audio for speech-to-text, and a HIPAA plan requirement for recordings

As data: https://www.anchorterminal.com/api/v1/use-cases/voice-calling.record-and-retrieve.json, or the MCP tool get_use_case with id voice-calling.record-and-retrieve. Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person.

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp
