# What changed in assemblyai

> 65 changes in assemblyai 0.53 to 1 from its official release notes, upgrade guides and advisories: 7 breaking, 44 new APIs and options. Each one cited. Sources checked 2026-11-12. HTML: https://unstale.sh/pypi/assemblyai/

To see which of these apply to a repository, run `npx @unstale/cli init` in it.

## assemblyai 1

### Breaking changes

- The live connection becomes the only connection the sync client opens. (1.5.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.2)
- The LeMUR API and all aai.Lemur* names are removed; send transcript text to an LLM or use aai.LLMGateway instead. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)
- The extras install option is removed; installing with this package extra fails. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md), [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.0.0)
- The extras module, including MicrophoneStream, is removed; capture audio with another library and pass PCM to the streaming client. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md), [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.0.0)
- LeMUR support was removed from the SDK. (1.0.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.0.0)
- RealTimeTranscriber was introduced as part of a transcriber rename. (1.0.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.0.0)

### New APIs and options

- The streaming API accepts the veterinary-v1 domain. (1.9.1) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.9.1)
- Adds a Text-to-Speech HTTP client for developers to use. (1.8.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.8.2)
- Adds a WebSocket session client for developers to use. (1.8.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.8.2)
- DictationConfig gains rewrite_mode, allowing developers to configure dictation rewrite behavior. (1.6.4) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.6.4)
- Adds an LLM Gateway module. (1.6.1) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.6.1)
- Adds the speaker_labels_revision_interval_ms parameter to streaming. (1.6.1) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.6.1)
- Adds the acknowledge_silence parameter for streaming. (1.5.5) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.5)
- Adds the Silence event for streaming. (1.5.5) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.5)
- Adds TranscriptConfig to improve the Speech Understanding UX. (1.5.4) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.4)
- Adds the universal-3-6 speech model for streaming. (1.5.4) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.4)
- Adds the Dictation API, with live upload as its only connection. (1.5.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.2)
- Adds stt_prompt to DictationConfig. (1.5.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.2)
- Adds push-style live sessions via open_live(). (1.5.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.2)
- Adds transcribe_stream() to stream the sync upload. (1.5.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.2)
- Adds a localization option to language_detection_options. (1.3.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.3.0)
- Adds short_file_diarization_method to SpeakerOptions. (1.3.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.3.0)
- Adds the universal-3-6-pro speech model for use with the SDK. (1.1.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.1.0)
- Client, transcriber, and streaming constructors now accept api_key=. (1.0.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.0.0)
- A versioned subpackage provides the preferred import path for prerecorded transcription APIs. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)
- A versioned subpackage provides the preferred import path for streaming APIs. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)
- A versioned subpackage provides the preferred import path for sync transcription APIs. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)
- Added AsyncSyncTranscriber for asyncio callers. (1.0.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.0.0)
- Async transcribers can be used as async context managers to release their HTTP connection pools; aclose() is the explicit equivalent. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)
- The streaming surface gains RealTime* names as the preferred names for the former Streaming* classes; old names remain bound to the same objects. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)
- SyncTranscriber gains warm() to establish a connection before transcription; AsyncSyncTranscriber has an awaitable equivalent. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)
- The transcribe methods accept poll_timeout to bound polling and prevent a stuck job from hanging a request handler. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)
- Transcribers accept a per-client API key, avoiding the need to configure process-wide state; when combined with client=, the key takes precedence. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)

### Behavior and default changes

- TTS sessions share one TLS context and HTTP connections are kept open for 120 seconds. (1.9.1) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.9.1)
- Raises the sync and dictation configuration caps to match the service. (1.5.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.5.2)
- Local files can now be passed as pathlib.Path objects directly, without converting them to strings. (1.0.0) — [upgrade guide](https://github.com/AssemblyAI/assemblyai-python-sdk/blob/HEAD/MIGRATION.md)

### Tooling

- Ships a PEP 561 py.typed marker so type checkers can recognize the package's type information. (1.3.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.3.0)

### Fixes and other changes

- Calling cancel() ends a reader after the request is written. (1.8.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.8.2)
- Typing was added or fixed for LLMGW Speech Understanding Summarization requests through the TranscriptConfig API. (1.6.5) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.6.5)
- Non-negative validation constraints were added to a RealTimeTr… component. (1.6.4) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.6.4)
- TLS reads are held until the handshake request is written, fixing sync client streaming behavior. (1.6.4) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.6.4)
- Guards model_config so the pydantic v1 fallback keeps its config. (1.3.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.3.0)
- Fixes the scan so it no longer drops later channels. (1.3.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/1.3.0)

## assemblyai 0.65

### Breaking changes

- Prerecorded transcription was moved into prerecorded/v2, changing where developers access it. (0.65.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.65.0)

### New APIs and options

- Adds AsyncTranscriber for callers using asyncio. (0.65.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.65.0)

## assemblyai 0.64

### New APIs and options

- MicrophoneStream adds pause() and resume() methods. (0.64.33) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.33)
- Sentences and paragraphs responses now include speech_model_used. (0.64.33) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.33)
- The transcribe type hints now accept bytes. (0.64.33) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.33)
- Opus is now supported for streaming encodings. (0.64.25) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.25)
- Synchronous connection pre-warming is now available. (0.64.25) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.25)
- Adds streaming sync API support. (0.64.9) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.9)
- Adds u3-rt-pro-beta-1 as a streaming speech model. (0.64.9) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.9)
- Added the turn_left_pad_ms parameter; the notes do not specify its behavior or purpose. (0.64.3) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.3)
- You can now specify language_detection_model inside language_detection_options. (0.64.2) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.2)
- Adds an Interruption Delay parameter for FAST TTFT. (0.64.1) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.1)

### Behavior and default changes

- language_codes can now be updated mid-stream. (0.64.25) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.25)
- Makes speech_model optional. (0.64.9) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.9)

### Fixes and other changes

- AutohighlightResult.rank is now Optional to tolerate a missing field. (0.64.33) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.33)
- The upload type hints now accept bytes. (0.64.33) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.64.33)

## assemblyai 0.59

### New APIs and options

- Adds domain support for medical mode, allowing developers to use the medical domain. (0.59.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.59.0)

## assemblyai 0.58

### New APIs and options

- Adds support for redact_pii_audio_options. (0.58.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.58.0)

## assemblyai 0.54

### New APIs and options

- Adds a Speech Started event for u3 pro streaming. (0.54.1) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.54.1)
- Added streaming diarization. (0.54.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.54.0)
- Added whisper rt. (0.54.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.54.0)

## assemblyai 0.53

### New APIs and options

- Added support for u3pro rt. (0.53.0) — [release notes](https://github.com/AssemblyAI/assemblyai-python-sdk/releases/tag/0.53.0)