diff --git a/CHANGELOG.md b/CHANGELOG.md
index b96eb9876..30acbdbd6 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -7,6 +7,308 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 <!-- towncrier release notes start -->
 
+## [0.0.108] - 2026-03-27
+
+### Added
+
+- Added `SarvamLLMService` with support for `sarvam-30b`, `sarvam-30b-16k`,
+  `sarvam-105b` and `sarvam-105b-32k`.
+  (PR [#3978](https://github.com/pipecat-ai/pipecat/pull/3978))
+
+- Added `on_turn_context_created(context_id)` hook to `TTSService`. Override
+  this to perform provider-specific setup (e.g. eagerly opening a server-side
+  context) before text starts flowing. Called each time a new turn context ID
+  is created.
+  (PR [#4013](https://github.com/pipecat-ai/pipecat/pull/4013))
+
+- Added `XAIHttpTTSService` for text-to-speech using xAI's HTTP TTS API.
+  (PR [#4031](https://github.com/pipecat-ai/pipecat/pull/4031))
+
+- Added support for "developer" role messages in conversation context across
+  all LLM adapters. For non-OpenAI services (Anthropic, Google, AWS Bedrock),
+  "developer" messages are converted to "user" messages (use
+  `system_instruction` to set the system instruction). For OpenAI services,
+  "developer" messages pass through in conversation history. For the Responses
+  API, they are kept as "developer" role (matching the existing "system" →
+  "developer" conversion).
+  (PR [#4089](https://github.com/pipecat-ai/pipecat/pull/4089))
+
+- Added `SmallestTTSService`, a WebSocket-based TTS service integration with
+  Smallest AI's Waves API. Supports the Lightning v2 and v3.1 models with
+  configurable voice, language, speed, consistency, similarity, and enhancement
+  settings.
+  (PR [#4092](https://github.com/pipecat-ai/pipecat/pull/4092))
+
+- Added warnings in turn stop strategies when `VADParams.stop_secs` differs
+  from the recommended default (0.2s) or when `stop_secs >= STT p99 latency`,
+  which collapses the STT wait timeout to 0s and may cause delayed turn
+  detection. The warnings guide developers to re-run the
+  [stt-benchmark](https://github.com/pipecat-ai/stt-benchmark) with their VAD
+  settings.
+  (PR [#4115](https://github.com/pipecat-ai/pipecat/pull/4115))
+
+- Added `domain` parameter to `AssemblyAISTTSettings` for specialized
+  recognition modes such as Medical Mode (`domain="medical-v1"`).
+  (PR [#4117](https://github.com/pipecat-ai/pipecat/pull/4117))
+
+- Added `NovitaLLMService` for using Novita AI's LLM models via their
+  OpenAI-compatible API.
+  (PR [#4119](https://github.com/pipecat-ai/pipecat/pull/4119))
+
+- Added `cleanup()` method to `VADAnalyzer` and `VADController` so VAD analyzer
+  resources are properly released when no longer needed. Custom `VADAnalyzer`
+  subclasses can override `cleanup()` to free any held resources.
+  (PR [#4120](https://github.com/pipecat-ai/pipecat/pull/4120))
+
+- Added `on_end_of_turn` event handler to `AssemblyAISTTService`. This fires
+  after the final transcript is pushed, providing a reliable hook for
+  end-of-turn logic that doesn't race with `TranscriptionFrame`. Works in both
+  Pipecat and AssemblyAI turn detection modes.
+  (PR [#4128](https://github.com/pipecat-ai/pipecat/pull/4128))
+
+- Added `DeepgramFluxSageMakerSTTService` for running Deepgram Flux
+  speech-to-text on AWS SageMaker endpoints.  Use with
+  `ExternalUserTurnStrategies` to take advantage of Flux's turn detection.
+  (PR [#4143](https://github.com/pipecat-ai/pipecat/pull/4143))
+
+- Added `Mem0MemoryService.get_memories()` convenience method for retrieving
+  all stored memories outside the pipeline (e.g. to build a personalized
+  greeting at connection time). This avoids the need to manually handle client
+  type branching, filter construction, and async wrapping.
+  (PR [#4156](https://github.com/pipecat-ai/pipecat/pull/4156))
+
+### Changed
+
+- Added context prewarming path for `InworldTTSService` to improve first audio
+  latency.
+  (PR [#4013](https://github.com/pipecat-ai/pipecat/pull/4013))
+
+- Added `KrispVivaVadAnalyzer` for Voice Activity Detection using the Krisp
+  VIVA SDK (requires `krisp_audio`).
+  (PR [#4022](https://github.com/pipecat-ai/pipecat/pull/4022))
+
+- Modified `InworldTTSService` to close context at end of turn instead of
+  relying on idle timeout.
+  (PR [#4028](https://github.com/pipecat-ai/pipecat/pull/4028))
+
+- Added Gemini 3 support to the Gemini Live service.
+  (PR [#4078](https://github.com/pipecat-ai/pipecat/pull/4078))
+
+- `TTSService`: the default `stop_frame_timeout_s` (idle time before an
+  automatic `TTSStoppedFrame` is pushed when `push_stop_frames=True`) has
+  changed from `2.0` to `3.0` seconds.
+  (PR [#4084](https://github.com/pipecat-ai/pipecat/pull/4084))
+
+- ⚠️ `GeminiLLMAdapter` now only treats `messages[0]` as the initial system
+  message, matching all other adapters. Previously it searched for the first
+  "system" message anywhere in the conversation history. A "system" message
+  appearing later in the list will now be converted to "user" instead of being
+  extracted as the system instruction.
+  (PR [#4089](https://github.com/pipecat-ai/pipecat/pull/4089))
+
+- Fixed `InworldTtsService` to fallback to full text when TTS timestamps are
+  not received.
+  (PR [#4113](https://github.com/pipecat-ai/pipecat/pull/4113))
+
+- ⚠️ Realtime services (Gemini Live, OpenAI Realtime, Grok Realtime, Nova
+  Sonic) now prefer `system_instruction` from service settings over an initial
+  system message in the LLM context, matching the behavior of non-realtime
+  services. Previously, context-provided system instructions took precedence. A
+  warning is now logged when both are set.
+  (PR [#4130](https://github.com/pipecat-ai/pipecat/pull/4130))
+
+- Bumped `nvidia-riva-client` minimum version to `>=2.25.1`.
+  (PR [#4136](https://github.com/pipecat-ai/pipecat/pull/4136))
+
+- Upgraded `protobuf` from 5.x to 6.x (`>=6.31.1,<7`).
+  (PR [#4136](https://github.com/pipecat-ai/pipecat/pull/4136))
+
+- Unrecognized language strings (e.g. Deepgram's `"multi"`) no longer produce a
+  warning at startup. The log message has been downgraded to debug level since
+  these are valid service-specific values that are passed through correctly.
+  (PR [#4137](https://github.com/pipecat-ai/pipecat/pull/4137))
+
+- `GrokLLMService` and `GrokRealtimeLLMService` now live in the
+  `pipecat.services.xai` module alongside `XAIHttpTTSService`, since all three
+  use the same xAI API. Update imports from `pipecat.services.grok.*` to
+  `pipecat.services.xai.*` (e.g. `from pipecat.services.xai.llm import
+  GrokLLMService`).
+  (PR [#4142](https://github.com/pipecat-ai/pipecat/pull/4142))
+
+- ⚠️ Bumped `mem0ai` dependency from `~=0.1.94` to `>=1.0.8,<2`. Users of the
+  `mem0` extra will need to update their mem0ai package.
+  (PR [#4156](https://github.com/pipecat-ai/pipecat/pull/4156))
+
+### Deprecated
+
+- `pipecat.services.grok.llm`, `pipecat.services.grok.realtime.llm`, and
+  `pipecat.services.grok.realtime.events` are deprecated. The old import paths
+  still work but emit a `DeprecationWarning`; use `pipecat.services.xai.llm`,
+  `pipecat.services.xai.realtime.llm`, and
+  `pipecat.services.xai.realtime.events` instead.
+  (PR [#4142](https://github.com/pipecat-ai/pipecat/pull/4142))
+
+### Removed
+
+- ⚠️ `TTSService.add_word_timestamps()` no longer supports the `"Reset"` and
+  `"TTSStoppedFrame"` sentinel strings. If you have a custom TTS service that
+  called `await self.add_word_timestamps([("Reset", 0)])` or `await
+  self.add_word_timestamps([("TTSStoppedFrame", 0), ("Reset", 0)], ctx_id)`,
+  replace them with `await self.append_to_audio_context(ctx_id,
+  TTSStoppedFrame(context_id=ctx_id))` and let `_handle_audio_context` manage
+  the word-timestamp reset automatically.
+  (PR [#4145](https://github.com/pipecat-ai/pipecat/pull/4145))
+
+- Removed `SambaNovaSTTService`. SambaNova no longer offers speech-to-text
+  audio models. Use another STT provider instead.
+  (PR [#4154](https://github.com/pipecat-ai/pipecat/pull/4154))
+
+### Fixed
+
+- Fixed Gemini Live (`GoogleGeminiLiveLLMService`) not honoring
+  `settings.system_instruction`. The system instruction was being read from a
+  deprecated constructor parameter instead of the settings object, causing it
+  to be silently ignored.
+  (PR [#4089](https://github.com/pipecat-ai/pipecat/pull/4089))
+
+- Fixed `AWSBedrockLLMAdapter` sending an empty message list to the API when
+  the only message in context was a system message. The lone system message is
+  now converted to "user" role instead of being extracted, matching the
+  existing Anthropic adapter behavior.
+  (PR [#4089](https://github.com/pipecat-ai/pipecat/pull/4089))
+
+- Fixed Gemini Live pipeline hanging indefinitely when an `EndFrame` was
+  deferred while waiting for the bot to finish responding and `turn_complete`
+  never arrived. As a possible root-cause fix, `turn_complete` messages are now
+  handled even if they lack `usage_metadata`. As a fallback, the deferred
+  `EndFrame` now has a 30-second safety timeout.
+  (PR [#4125](https://github.com/pipecat-ai/pipecat/pull/4125))
+
+- Fixed ElevenLabs WebSocket disconnections (1008 "Maximum simultaneous
+  contexts exceeded") caused by rapid user interruptions. When interruptions
+  arrived before any TTS text was generated, phantom contexts were created on
+  the ElevenLabs server that were never closed, eventually exceeding the
+  5-context limit.
+  (PR [#4126](https://github.com/pipecat-ai/pipecat/pull/4126))
+
+- Fixed the final sentence being dropped from the conversation context when
+  using RTVI text input with non-word-timestamp TTS services. The
+  `LLMFullResponseEndFrame` was racing ahead of the last `TTSTextFrame`,
+  causing the `LLMAssistantAggregator` to finalize the context before the final
+  sentence arrived.
+  (PR [#4127](https://github.com/pipecat-ai/pipecat/pull/4127))
+
+- Fixed audio crackling and popping in recordings when both user and bot are
+  speaking. `AudioBufferProcessor` no longer injects silence into a track's
+  buffer while that track is actively producing audio, preventing mid-utterance
+  interruptions in the recorded output.
+  (PR [#4135](https://github.com/pipecat-ai/pipecat/pull/4135))
+
+- Fixed websocket TTS word timestamps so interrupted contexts cannot leak stale
+  words or backward PTS values into later turns.
+  (PR [#4145](https://github.com/pipecat-ai/pipecat/pull/4145))
+
+- Fixed a race condition in `InterruptibleTTSService` where, if `run_tts` had
+  been invoked but `BotStartedSpeakingFrame` had not yet been received, a user
+  interruption could allow stale audio to leak through.
+  (PR [#4145](https://github.com/pipecat-ai/pipecat/pull/4145))
+
+- Fixed Gemini Live local VAD mode (`GeminiVADParams(disabled=True)` with
+  external VAD) not working. The bot now correctly detects user speech and
+  signals turn boundaries to the Gemini API.
+  (PR [#4146](https://github.com/pipecat-ai/pipecat/pull/4146))
+
+- Fixed Gemini Live message handling to process all `server_content` fields
+  independently. Gemini 3.x can bundle multiple fields (e.g. `model_turn` and
+  `output_transcription`) on the same message, but the previous `elif` chain
+  only processed the first match, silently dropping the rest.
+  (PR [#4147](https://github.com/pipecat-ai/pipecat/pull/4147))
+
+- Fixed `ServiceSwitcher` with `ServiceSwitcherStrategyFailover` incorrectly
+  triggering failover when `ErrorFrame`s from other pipeline stages (e.g. TTS)
+  propagated upstream through the switcher. Previously, any non-fatal error
+  passing through would be misattributed to the active service and trigger an
+  unwanted service switch. Now only errors originating from the switcher's own
+  managed services trigger failover.
+  (PR [#4149](https://github.com/pipecat-ai/pipecat/pull/4149))
+
+- Fixed `LiveKitOutputTransport` not clearing the `rtc.AudioSource` internal
+  buffer on interruption, causing the bot to continue speaking for several
+  seconds after being interrupted.
+  (PR [#4151](https://github.com/pipecat-ai/pipecat/pull/4151))
+
+- Fixed a crash in OpenAI LLM processing when the provider returns
+  `chunk.choices[0].delta.audio = None`, which caused `'NoneType' object has no
+  attribute 'get'` errors during audio transcript handling.
+  (PR [#4152](https://github.com/pipecat-ai/pipecat/pull/4152))
+
+- Fixed error floods in `DeepgramSTTService` when the WebSocket connection
+  drops. With Deepgram SDK 6.x, `send_media()` raises exceptions on a dead
+  connection instead of silently failing, causing every queued audio frame to
+  log an error. Now `send_media()` failures are caught gracefully — a single
+  warning is logged and audio frames are skipped until the existing
+  reconnection logic restores the connection.
+  (PR [#4153](https://github.com/pipecat-ai/pipecat/pull/4153))
+
+- `Mem0MemoryService` no longer blocks the event loop during memory storage and
+  retrieval. All Mem0 API calls now run in a background thread, and message
+  storage is fire-and-forget so it doesn't delay downstream processing.
+  (PR [#4156](https://github.com/pipecat-ai/pipecat/pull/4156))
+
+- Fixed `Mem0MemoryService` failing to store messages when the context
+  contained system or developer role messages. The Mem0 API only accepts user
+  and assistant roles, so other roles are now filtered out before storing.
+  (PR [#4156](https://github.com/pipecat-ai/pipecat/pull/4156))
+
+- Added missing `on_dtmf_event` callback to `LemonSliceTransportClient.setup()`
+  `DailyCallbacks` construction, fixing a `ValidationError` at pipeline setup
+  time.
+  (PR [#4161](https://github.com/pipecat-ai/pipecat/pull/4161))
+
+- Fixed an issue in `InworldTTSService` where, in cases of fast interruption,
+  we would continue receiving audio from the previous context.
+  (PR [#4167](https://github.com/pipecat-ai/pipecat/pull/4167))
+
+- Fixed a word timestamp interleaving issue in `InworldTTSService` when
+  processing multiple sentences.
+  (PR [#4167](https://github.com/pipecat-ai/pipecat/pull/4167))
+
+- Fixed duplicate `TTSStoppedFrame` being pushed in TTS services using
+  `push_stop_frames=True`. When the stop-frame timeout fired, a second
+  `TTSStoppedFrame` could be pushed after the normal one at context completion.
+  (PR [#4172](https://github.com/pipecat-ai/pipecat/pull/4172))
+
+- ⚠️ Fixed `DeepgramSTTService` compatibility with deepgram-sdk 6.1.0. The SDK
+  now requires explicit message objects for `send_keep_alive()`,
+  `send_close_stream()`, and `send_finalize()`. The minimum deepgram-sdk
+  version is now 6.1.0.
+  (PR [#4174](https://github.com/pipecat-ai/pipecat/pull/4174))
+
+- Fixed RTVI events not being delivered to clients when using WebSocket
+  transports. `ProtobufFrameSerializer` now sets `ignore_rtvi_messages=False`
+  by default.
+  (PR [#4176](https://github.com/pipecat-ai/pipecat/pull/4176))
+
+- Fixed a timing issue where turn detection timer tasks (idle controller,
+  speech timeout, turn analyzer, and turn completion) could miss their first
+  tick because the newly created asyncio task was not yet scheduled when the
+  caller continued.
+  (PR [#4183](https://github.com/pipecat-ai/pipecat/pull/4183))
+
+- Fixed `FastAPIWebsocketTransport` intermittently hanging on shutdown when the
+  remote side (e.g. Twilio) disconnects while audio is being sent. A race
+  condition between the send and receive paths could cause the
+  `on_client_disconnected` callback to be skipped, leaving the pipeline waiting
+  for a disconnect signal that never came.
+  (PR [#4186](https://github.com/pipecat-ai/pipecat/pull/4186))
+
+### Performance
+
+- `RimeTTSService` now handles Rime's `done` WebSocket message to complete
+  audio contexts immediately, eliminating the 3-second idle timeout that
+  previously added latency at the end of each utterance.
+  (PR [#4172](https://github.com/pipecat-ai/pipecat/pull/4172))
+
 ## [0.0.107] - 2026-03-23
 
 ### Added
diff --git a/changelog/3978.added.md b/changelog/3978.added.md
deleted file mode 100644
index bb80b10c0..000000000
--- a/changelog/3978.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `SarvamLLMService` with support for `sarvam-30b`, `sarvam-30b-16k`, `sarvam-105b` and `sarvam-105b-32k`
diff --git a/changelog/4013.added.md b/changelog/4013.added.md
deleted file mode 100644
index 1be433cb3..000000000
--- a/changelog/4013.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `on_turn_context_created(context_id)` hook to `TTSService`. Override this to perform provider-specific setup (e.g. eagerly opening a server-side context) before text starts flowing. Called each time a new turn context ID is created.
\ No newline at end of file
diff --git a/changelog/4013.changed.md b/changelog/4013.changed.md
deleted file mode 100644
index 4f8a37680..000000000
--- a/changelog/4013.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added context prewarming path for `InworldTTSService` to improve first audio latency
\ No newline at end of file
diff --git a/changelog/4022.changed.md b/changelog/4022.changed.md
deleted file mode 100644
index 7669f0810..000000000
--- a/changelog/4022.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `KrispVivaVadAnalyzer` for Voice Activity Detection using the Krisp VIVA SDK (requires `krisp_audio`).
diff --git a/changelog/4028.changed.md b/changelog/4028.changed.md
deleted file mode 100644
index 56d5d2f17..000000000
--- a/changelog/4028.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Modeified `InworldTTSService` to close context at end of turn instead of relying on idle timeout
diff --git a/changelog/4031.added.md b/changelog/4031.added.md
deleted file mode 100644
index 0c228a24a..000000000
--- a/changelog/4031.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `XAIHttpTTSService` for text-to-speech using xAI's HTTP TTS API.
diff --git a/changelog/4078.changed.md b/changelog/4078.changed.md
deleted file mode 100644
index c0f4fb0a9..000000000
--- a/changelog/4078.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added Gemini 3 support to the Gemini Live service.
diff --git a/changelog/4084.changed.md b/changelog/4084.changed.md
deleted file mode 100644
index 63fddf56b..000000000
--- a/changelog/4084.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- `TTSService`: the default `stop_frame_timeout_s` (idle time before an automatic `TTSStoppedFrame` is pushed when `push_stop_frames=True`) has changed from `2.0` to `3.0` seconds.
diff --git a/changelog/4089.added.md b/changelog/4089.added.md
deleted file mode 100644
index b1ec724f5..000000000
--- a/changelog/4089.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added support for "developer" role messages in conversation context across all LLM adapters. For non-OpenAI services (Anthropic, Google, AWS Bedrock), "developer" messages are converted to "user" messages (use `system_instruction` to set the system instruction). For OpenAI services, "developer" messages pass through in conversation history. For the Responses API, they are kept as "developer" role (matching the existing "system" → "developer" conversion).
diff --git a/changelog/4089.changed.md b/changelog/4089.changed.md
deleted file mode 100644
index f44a7b772..000000000
--- a/changelog/4089.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- ⚠️ `GeminiLLMAdapter` now only treats `messages[0]` as the initial system message, matching all other adapters. Previously it searched for the first "system" message anywhere in the conversation history. A "system" message appearing later in the list will now be converted to "user" instead of being extracted as the system instruction.
diff --git a/changelog/4089.fixed.2.md b/changelog/4089.fixed.2.md
deleted file mode 100644
index 43f302096..000000000
--- a/changelog/4089.fixed.2.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed Gemini Live (`GoogleGeminiLiveLLMService`) not honoring `settings.system_instruction`. The system instruction was being read from a deprecated constructor parameter instead of the settings object, causing it to be silently ignored.
diff --git a/changelog/4089.fixed.md b/changelog/4089.fixed.md
deleted file mode 100644
index 03cadf4d2..000000000
--- a/changelog/4089.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed `AWSBedrockLLMAdapter` sending an empty message list to the API when the only message in context was a system message. The lone system message is now converted to "user" role instead of being extracted, matching the existing Anthropic adapter behavior.
diff --git a/changelog/4092.added.md b/changelog/4092.added.md
deleted file mode 100644
index aa4dcf269..000000000
--- a/changelog/4092.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `SmallestTTSService`, a WebSocket-based TTS service integration with Smallest AI's Waves API. Supports the Lightning v2 and v3.1 models with configurable voice, language, speed, consistency, similarity, and enhancement settings.
diff --git a/changelog/4113.changed.md b/changelog/4113.changed.md
deleted file mode 100644
index 7a30f71d5..000000000
--- a/changelog/4113.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed `InworldTtsService` to fallback to full text when TTS timestamps are not received
diff --git a/changelog/4115.added.md b/changelog/4115.added.md
deleted file mode 100644
index 4736dff5a..000000000
--- a/changelog/4115.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added warnings in turn stop strategies when `VADParams.stop_secs` differs from the recommended default (0.2s) or when `stop_secs >= STT p99 latency`, which collapses the STT wait timeout to 0s and may cause delayed turn detection. The warnings guide developers to re-run the [stt-benchmark](https://github.com/pipecat-ai/stt-benchmark) with their VAD settings.
diff --git a/changelog/4117.added.md b/changelog/4117.added.md
deleted file mode 100644
index b59eceb60..000000000
--- a/changelog/4117.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `domain` parameter to `AssemblyAISTTSettings` for specialized recognition modes such as Medical Mode (`domain="medical-v1"`).
diff --git a/changelog/4119.added.md b/changelog/4119.added.md
deleted file mode 100644
index 7f3003f3a..000000000
--- a/changelog/4119.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `NovitaLLMService` for using Novita AI's LLM models via their OpenAI-compatible API.
diff --git a/changelog/4120.added.md b/changelog/4120.added.md
deleted file mode 100644
index 71e514271..000000000
--- a/changelog/4120.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `cleanup()` method to `VADAnalyzer` and `VADController` so VAD analyzer resources are properly released when no longer needed. Custom `VADAnalyzer` subclasses can override `cleanup()` to free any held resources.
diff --git a/changelog/4125.fixed.md b/changelog/4125.fixed.md
deleted file mode 100644
index 7db696a39..000000000
--- a/changelog/4125.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed Gemini Live pipeline hanging indefinitely when an `EndFrame` was deferred while waiting for the bot to finish responding and `turn_complete` never arrived. As a possible root-cause fix, `turn_complete` messages are now handled even if they lack `usage_metadata`. As a fallback, the deferred `EndFrame` now has a 30-second safety timeout.
diff --git a/changelog/4126.fixed.md b/changelog/4126.fixed.md
deleted file mode 100644
index 205568bae..000000000
--- a/changelog/4126.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed ElevenLabs WebSocket disconnections (1008 "Maximum simultaneous contexts exceeded") caused by rapid user interruptions. When interruptions arrived before any TTS text was generated, phantom contexts were created on the ElevenLabs server that were never closed, eventually exceeding the 5-context limit.
diff --git a/changelog/4127.fixed.md b/changelog/4127.fixed.md
deleted file mode 100644
index 6735808d3..000000000
--- a/changelog/4127.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed the final sentence being dropped from the conversation context when using RTVI text input with non-word-timestamp TTS services. The `LLMFullResponseEndFrame` was racing ahead of the last `TTSTextFrame`, causing the `LLMAssistantAggregator` to finalize the context before the final sentence arrived.
diff --git a/changelog/4128.added.md b/changelog/4128.added.md
deleted file mode 100644
index 7cbeed91b..000000000
--- a/changelog/4128.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `on_end_of_turn` event handler to `AssemblyAISTTService`. This fires after the final transcript is pushed, providing a reliable hook for end-of-turn logic that doesn't race with `TranscriptionFrame`. Works in both Pipecat and AssemblyAI turn detection modes.
diff --git a/changelog/4130.changed.md b/changelog/4130.changed.md
deleted file mode 100644
index 66b38ffb7..000000000
--- a/changelog/4130.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- ⚠️ Realtime services (Gemini Live, OpenAI Realtime, Grok Realtime, Nova Sonic) now prefer `system_instruction` from service settings over an initial system message in the LLM context, matching the behavior of non-realtime services. Previously, context-provided system instructions took precedence. A warning is now logged when both are set.
diff --git a/changelog/4135.fixed.md b/changelog/4135.fixed.md
deleted file mode 100644
index 92f57ce83..000000000
--- a/changelog/4135.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed audio crackling and popping in recordings when both user and bot are speaking. `AudioBufferProcessor` no longer injects silence into a track's buffer while that track is actively producing audio, preventing mid-utterance interruptions in the recorded output.
diff --git a/changelog/4136.changed.2.md b/changelog/4136.changed.2.md
deleted file mode 100644
index 57160ea62..000000000
--- a/changelog/4136.changed.2.md
+++ /dev/null
@@ -1 +0,0 @@
-- Bumped `nvidia-riva-client` minimum version to `>=2.25.1`.
diff --git a/changelog/4136.changed.md b/changelog/4136.changed.md
deleted file mode 100644
index 61e7fc242..000000000
--- a/changelog/4136.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Upgraded `protobuf` from 5.x to 6.x (`>=6.31.1,<7`).
diff --git a/changelog/4137.changed.md b/changelog/4137.changed.md
deleted file mode 100644
index a0b9f2aae..000000000
--- a/changelog/4137.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Unrecognized language strings (e.g. Deepgram's `"multi"`) no longer produce a warning at startup. The log message has been downgraded to debug level since these are valid service-specific values that are passed through correctly.
diff --git a/changelog/4142.changed.md b/changelog/4142.changed.md
deleted file mode 100644
index 06fa50f74..000000000
--- a/changelog/4142.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- `GrokLLMService` and `GrokRealtimeLLMService` now live in the `pipecat.services.xai` module alongside `XAIHttpTTSService`, since all three use the same xAI API. Update imports from `pipecat.services.grok.*` to `pipecat.services.xai.*` (e.g. `from pipecat.services.xai.llm import GrokLLMService`).
diff --git a/changelog/4142.deprecated.md b/changelog/4142.deprecated.md
deleted file mode 100644
index a17444edd..000000000
--- a/changelog/4142.deprecated.md
+++ /dev/null
@@ -1 +0,0 @@
-- `pipecat.services.grok.llm`, `pipecat.services.grok.realtime.llm`, and `pipecat.services.grok.realtime.events` are deprecated. The old import paths still work but emit a `DeprecationWarning`; use `pipecat.services.xai.llm`, `pipecat.services.xai.realtime.llm`, and `pipecat.services.xai.realtime.events` instead.
diff --git a/changelog/4143.added.md b/changelog/4143.added.md
deleted file mode 100644
index 0f26b68b0..000000000
--- a/changelog/4143.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `DeepgramFluxSageMakerSTTService` for running Deepgram Flux speech-to-text on AWS SageMaker endpoints.  Use with `ExternalUserTurnStrategies` to take advantage of Flux's turn detection.
diff --git a/changelog/4145.fixed.2.md b/changelog/4145.fixed.2.md
deleted file mode 100644
index 4fceb9bb1..000000000
--- a/changelog/4145.fixed.2.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed websocket TTS word timestamps so interrupted contexts cannot leak stale words or backward PTS values into later turns.
\ No newline at end of file
diff --git a/changelog/4145.fixed.md b/changelog/4145.fixed.md
deleted file mode 100644
index 831e1fcd7..000000000
--- a/changelog/4145.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed a race condition in `InterruptibleTTSService` where, if `run_tts` had been invoked but `BotStartedSpeakingFrame` had not yet been received, a user interruption could allow stale audio to leak through.
\ No newline at end of file
diff --git a/changelog/4145.removed.md b/changelog/4145.removed.md
deleted file mode 100644
index ac780c1dc..000000000
--- a/changelog/4145.removed.md
+++ /dev/null
@@ -1 +0,0 @@
-- ⚠️ `TTSService.add_word_timestamps()` no longer supports the `"Reset"` and `"TTSStoppedFrame"` sentinel strings. If you have a custom TTS service that called `await self.add_word_timestamps([("Reset", 0)])` or `await self.add_word_timestamps([("TTSStoppedFrame", 0), ("Reset", 0)], ctx_id)`, replace them with `await self.append_to_audio_context(ctx_id, TTSStoppedFrame(context_id=ctx_id))` and let `_handle_audio_context` manage the word-timestamp reset automatically.
diff --git a/changelog/4146.fixed.md b/changelog/4146.fixed.md
deleted file mode 100644
index db58dad06..000000000
--- a/changelog/4146.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed Gemini Live local VAD mode (`GeminiVADParams(disabled=True)` with external VAD) not working. The bot now correctly detects user speech and signals turn boundaries to the Gemini API.
diff --git a/changelog/4147.fixed.md b/changelog/4147.fixed.md
deleted file mode 100644
index 985990c4d..000000000
--- a/changelog/4147.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed Gemini Live message handling to process all `server_content` fields independently. Gemini 3.x can bundle multiple fields (e.g. `model_turn` and `output_transcription`) on the same message, but the previous `elif` chain only processed the first match, silently dropping the rest.
diff --git a/changelog/4149.fixed.md b/changelog/4149.fixed.md
deleted file mode 100644
index db4167633..000000000
--- a/changelog/4149.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed `ServiceSwitcher` with `ServiceSwitcherStrategyFailover` incorrectly triggering failover when `ErrorFrame`s from other pipeline stages (e.g. TTS) propagated upstream through the switcher. Previously, any non-fatal error passing through would be misattributed to the active service and trigger an unwanted service switch. Now only errors originating from the switcher's own managed services trigger failover.
diff --git a/changelog/4151.fixed.md b/changelog/4151.fixed.md
deleted file mode 100644
index 0bcd55194..000000000
--- a/changelog/4151.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed `LiveKitOutputTransport` not clearing the `rtc.AudioSource` internal buffer on interruption, causing the bot to continue speaking for several seconds after being interrupted.
diff --git a/changelog/4152.fixed.md b/changelog/4152.fixed.md
deleted file mode 100644
index e752bb1ac..000000000
--- a/changelog/4152.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed a crash in OpenAI LLM processing when the provider returns `chunk.choices[0].delta.audio = None`, which caused `'NoneType' object has no attribute 'get'` errors during audio transcript handling.
diff --git a/changelog/4153.fixed.md b/changelog/4153.fixed.md
deleted file mode 100644
index 1e3dbebd0..000000000
--- a/changelog/4153.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed error floods in `DeepgramSTTService` when the WebSocket connection drops. With Deepgram SDK 6.x, `send_media()` raises exceptions on a dead connection instead of silently failing, causing every queued audio frame to log an error. Now `send_media()` failures are caught gracefully — a single warning is logged and audio frames are skipped until the existing reconnection logic restores the connection.
diff --git a/changelog/4154.removed.md b/changelog/4154.removed.md
deleted file mode 100644
index 4a778cc48..000000000
--- a/changelog/4154.removed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Removed `SambaNovaSTTService`. SambaNova no longer offers speech-to-text audio models. Use another STT provider instead.
diff --git a/changelog/4156.added.md b/changelog/4156.added.md
deleted file mode 100644
index bbf426d5f..000000000
--- a/changelog/4156.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `Mem0MemoryService.get_memories()` convenience method for retrieving all stored memories outside the pipeline (e.g. to build a personalized greeting at connection time). This avoids the need to manually handle client type branching, filter construction, and async wrapping.
diff --git a/changelog/4156.changed.md b/changelog/4156.changed.md
deleted file mode 100644
index d4b8a269a..000000000
--- a/changelog/4156.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- ⚠️ Bumped `mem0ai` dependency from `~=0.1.94` to `>=1.0.8,<2`. Users of the `mem0` extra will need to update their mem0ai package.
diff --git a/changelog/4156.fixed.2.md b/changelog/4156.fixed.2.md
deleted file mode 100644
index 1a6997d4f..000000000
--- a/changelog/4156.fixed.2.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed `Mem0MemoryService` failing to store messages when the context contained system or developer role messages. The Mem0 API only accepts user and assistant roles, so other roles are now filtered out before storing.
diff --git a/changelog/4156.fixed.md b/changelog/4156.fixed.md
deleted file mode 100644
index 13620f772..000000000
--- a/changelog/4156.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- `Mem0MemoryService` no longer blocks the event loop during memory storage and retrieval. All Mem0 API calls now run in a background thread, and message storage is fire-and-forget so it doesn't delay downstream processing.
diff --git a/changelog/4161.fixed.md b/changelog/4161.fixed.md
deleted file mode 100644
index f8dc3c670..000000000
--- a/changelog/4161.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added missing `on_dtmf_event` callback to `LemonSliceTransportClient.setup()` `DailyCallbacks` construction, fixing a `ValidationError` at pipeline setup time.
diff --git a/changelog/4167.fixed.2.md b/changelog/4167.fixed.2.md
deleted file mode 100644
index 6b894cc7f..000000000
--- a/changelog/4167.fixed.2.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed an issue in `InworldTTSService` where, in cases of fast interruption, we would continue receiving audio from the previous context.
\ No newline at end of file
diff --git a/changelog/4167.fixed.md b/changelog/4167.fixed.md
deleted file mode 100644
index 2784d4a41..000000000
--- a/changelog/4167.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed a word timestamp interleaving issue in `InworldTTSService` when processing multiple sentences.
diff --git a/changelog/4172.fixed.md b/changelog/4172.fixed.md
deleted file mode 100644
index 2b7aeceaf..000000000
--- a/changelog/4172.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed duplicate `TTSStoppedFrame` being pushed in TTS services using `push_stop_frames=True`. When the stop-frame timeout fired, a second `TTSStoppedFrame` could be pushed after the normal one at context completion.
diff --git a/changelog/4172.performance.md b/changelog/4172.performance.md
deleted file mode 100644
index 10fa51c66..000000000
--- a/changelog/4172.performance.md
+++ /dev/null
@@ -1 +0,0 @@
-- `RimeTTSService` now handles Rime's `done` WebSocket message to complete audio contexts immediately, eliminating the 3-second idle timeout that previously added latency at the end of each utterance.
diff --git a/changelog/4174.fixed.md b/changelog/4174.fixed.md
deleted file mode 100644
index 874f7e3c7..000000000
--- a/changelog/4174.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- ⚠️ Fixed `DeepgramSTTService` compatibility with deepgram-sdk 6.1.0. The SDK now requires explicit message objects for `send_keep_alive()`, `send_close_stream()`, and `send_finalize()`. The minimum deepgram-sdk version is now 6.1.0.
diff --git a/changelog/4176.fixed.md b/changelog/4176.fixed.md
deleted file mode 100644
index 90e9b6b88..000000000
--- a/changelog/4176.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed RTVI events not being delivered to clients when using WebSocket transports. `ProtobufFrameSerializer` now sets `ignore_rtvi_messages=False` by default.
diff --git a/changelog/4183.fixed.md b/changelog/4183.fixed.md
deleted file mode 100644
index a1f5ad4c8..000000000
--- a/changelog/4183.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed a timing issue where turn detection timer tasks (idle controller, speech timeout, turn analyzer, and turn completion) could miss their first tick because the newly created asyncio task was not yet scheduled when the caller continued.
diff --git a/changelog/4186.fixed.md b/changelog/4186.fixed.md
deleted file mode 100644
index 1ca3f1122..000000000
--- a/changelog/4186.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed `FastAPIWebsocketTransport` intermittently hanging on shutdown when the remote side (e.g. Twilio) disconnects while audio is being sent. A race condition between the send and receive paths could cause the `on_client_disconnected` callback to be skipped, leaving the pipeline waiting for a disconnect signal that never came.