chore(deps): update github actions

[renovate]

This PR contains the following updates:

Package	Type	Update	Change
actions/checkout (changelog)	action	digest	9c091bb → 3d3c42e
python	uses-with	minor	3.12 → 3.14
python	uses-with	minor	3.10 → 3.14
python	uses-with	minor	3.11 → 3.14

---

Warning

Some dependencies could not be looked up. Check the Dependency Dashboard for more information.

---

Release Notes

actions/python-versions (python)

v3.14.7: 3.14.7

Compare Source

Python 3.14.7

v3.14.6: 3.14.6

Compare Source

Python 3.14.6

v3.14.5: 3.14.5

Compare Source

Python 3.14.5

v3.14.4: 3.14.4

Compare Source

Python 3.14.4

v3.14.3: 3.14.3

Compare Source

Python 3.14.3

v3.14.2: 3.14.2

Compare Source

Python 3.14.2

v3.14.1: 3.14.1

Compare Source

Python 3.14.1

v3.14.0: 3.14.0

Compare Source

Python 3.14.0

v3.13.15: 3.13.15

Compare Source

Python 3.13.15

v3.13.14: 3.13.14

Compare Source

Python 3.13.14

v3.13.13: 3.13.13

Compare Source

Python 3.13.13

v3.13.12: 3.13.12

Compare Source

Python 3.13.12

v3.13.11: 3.13.11

Compare Source

Python 3.13.11

v3.13.10: 3.13.10

Compare Source

Python 3.13.10

v3.13.9: 3.13.9

Compare Source

Python 3.13.9

v3.13.8: 3.13.8

Compare Source

Python 3.13.8

v3.13.7: 3.13.7

Compare Source

Python 3.13.7

v3.13.6: 3.13.6

Compare Source

Python 3.13.6

v3.13.5: 3.13.5

Compare Source

Python 3.13.5

v3.13.4: 3.13.4

Compare Source

Python 3.13.4

v3.13.3: 3.13.3

Compare Source

Python 3.13.3

v3.13.2: 3.13.2

Compare Source

Python 3.13.2

v3.13.1: 3.13.1

Compare Source

Python 3.13.1

v3.13.0: 3.13.0

Compare Source

Python 3.13.0

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • "before 9am on monday"
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

👻 Immortal: This PR will be recreated if closed unmerged. Get config help if that's undesired.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.

[devin-ai-integration]

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no potential bugs to report.

View in Devin Review to see 2 additional findings.

[github-actions]

STT Test Results

Status: ✗ Some tests failed

Metric	Count
✓ Passed	20
✗ Failed	5
× Errors	0
→ Skipped	16
▣ Total	41
⏱ Duration	219.2s

Failed Tests

• tests.test_stt::test_stream[livekit.plugins.assemblyai]

  stt_factory = <function parameter_factory.<locals>.<lambda> at 0x7f3d9d2e2140>
  request = <FixtureRequest for <Coroutine test_stream[livekit.plugins.assemblyai]>>
  
      @pytest.mark.usefixtures("job_process")
      @pytest.mark.parametrize("stt_factory", STTs + STREAM_ONLY_STTs)
      async def test_stream(stt_factory: Callable[[], STT], request):
          sample_rate = SAMPLE_RATE
          plugin_id = request.node.callspec.id.split("-")[0]
          frames, transcript, _ = await make_test_speech(chunk_duration_ms=10, sample_rate=sample_rate)
    
          # TODO: differentiate missing key vs other errors
          try:
              stt_instance: STT = stt_factory()
          except ValueError as e:
              pytest.skip(f"{plugin_id}: {e}")
    
          async with stt_instance as stt:
              label = f"{stt.model}@{stt.provider}"
              if not stt.capabilities.streaming:
                  pytest.skip(f"{label} does not support streaming")
    
              for attempt in range(MAX_RETRIES):
                  try:
                      state = {"closing": False}
    
                      async def _stream_input(
                          frames: list[rtc.AudioFrame], stream: RecognizeStream, state: dict = state
                      ):
                          for frame in frames:
                              stream.push_frame(frame)
                              await asyncio.sleep(0.005)
    
                          stream.end_input()
                          state["closing"] = True
    
                      async def _stream_output(stream: RecognizeStream, state: dict = state):
                          text = ""
                          # make sure the events are sent in the right order
                          recv_start, recv_end = False, True
                          start_time = time.time()
                          got_final_transcript = False
                          sos_count, final_count = 0, 0
    
                          async for event in stream:
                              if even
  

• tests.test_stt::test_stream[livekit.plugins.fireworksai]

  self = <livekit.plugins.fireworksai.stt.SpeechStream object at 0x7f24146b11d0>
  
      async def __anext__(self) -> SpeechEvent:
          try:
  >           val = await self._event_aiter.__anext__()
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  E           StopAsyncIteration
  
  livekit-agents/livekit/agents/stt/stt.py:594: StopAsyncIteration
  
  During handling of the above exception, another exception occurred:
  
  stt_factory = <function parameter_factory.<locals>.<lambda> at 0x7f2415f4df30>
  request = <FixtureRequest for <Coroutine test_stream[livekit.plugins.fireworksai]>>
  
      @pytest.mark.usefixtures("job_process")
      @pytest.mark.parametrize("stt_factory", STTs + STREAM_ONLY_STTs)
      async def test_stream(stt_factory: Callable[[], STT], request):
          sample_rate = SAMPLE_RATE
          plugin_id = request.node.callspec.id.split("-")[0]
          frames, transcript, _ = await make_test_speech(chunk_duration_ms=10, sample_rate=sample_rate)
    
          # TODO: differentiate missing key vs other errors
          try:
              stt_instance: STT = stt_factory()
          except ValueError as e:
              pytest.skip(f"{plugin_id}: {e}")
    
          async with stt_instance as stt:
              label = f"{stt.model}@{stt.provider}"
              if not stt.capabilities.streaming:
                  pytest.skip(f"{label} does not support streaming")
    
              for attempt in range(MAX_RETRIES):
                  try:
                      state = {"closing": False}
    
                      async def _stream_input(
                          frames: list[rtc.AudioFrame], stream: RecognizeStream, state: dict = state
                      ):
                          for frame in frames:
                              stream.push_frame(frame)
                              await asyncio.sleep(0.005)
    
                          stream.end_input()
                          state["closing"] = True
    
                      async def _stream_output(stream: RecognizeStream, state: dict = state):
  

• tests.test_stt::test_stream[livekit.plugins.nvidia]

  stt_factory = <function parameter_factory.<locals>.<lambda> at 0x7f2415f4e1f0>
  request = <FixtureRequest for <Coroutine test_stream[livekit.plugins.nvidia]>>
  
      @pytest.mark.usefixtures("job_process")
      @pytest.mark.parametrize("stt_factory", STTs + STREAM_ONLY_STTs)
      async def test_stream(stt_factory: Callable[[], STT], request):
          sample_rate = SAMPLE_RATE
          plugin_id = request.node.callspec.id.split("-")[0]
          frames, transcript, _ = await make_test_speech(chunk_duration_ms=10, sample_rate=sample_rate)
    
          # TODO: differentiate missing key vs other errors
          try:
              stt_instance: STT = stt_factory()
          except ValueError as e:
              pytest.skip(f"{plugin_id}: {e}")
    
          async with stt_instance as stt:
              label = f"{stt.model}@{stt.provider}"
              if not stt.capabilities.streaming:
                  pytest.skip(f"{label} does not support streaming")
    
              for attempt in range(MAX_RETRIES):
                  try:
                      state = {"closing": False}
    
                      async def _stream_input(
                          frames: list[rtc.AudioFrame], stream: RecognizeStream, state: dict = state
                      ):
                          for frame in frames:
                              stream.push_frame(frame)
                              await asyncio.sleep(0.005)
    
                          stream.end_input()
                          state["closing"] = True
    
                      async def _stream_output(stream: RecognizeStream, state: dict = state):
                          text = ""
                          # make sure the events are sent in the right order
                          recv_start, recv_end = False, True
                          start_time = time.time()
                          got_final_transcript = False
                          sos_count, final_count = 0, 0
    
                          async for event in stream:
                              if event.ty
  

• tests.test_stt::test_stream[livekit.plugins.soniox]

  stt_factory = <function parameter_factory.<locals>.<lambda> at 0x7fdcae324b40>
  request = <FixtureRequest for <Coroutine test_stream[livekit.plugins.soniox]>>
  
      @pytest.mark.usefixtures("job_process")
      @pytest.mark.parametrize("stt_factory", STTs + STREAM_ONLY_STTs)
      async def test_stream(stt_factory: Callable[[], STT], request):
          sample_rate = SAMPLE_RATE
          plugin_id = request.node.callspec.id.split("-")[0]
          frames, transcript, _ = await make_test_speech(chunk_duration_ms=10, sample_rate=sample_rate)
    
          # TODO: differentiate missing key vs other errors
          try:
              stt_instance: STT = stt_factory()
          except ValueError as e:
              pytest.skip(f"{plugin_id}: {e}")
    
          async with stt_instance as stt:
              label = f"{stt.model}@{stt.provider}"
              if not stt.capabilities.streaming:
                  pytest.skip(f"{label} does not support streaming")
    
              for attempt in range(MAX_RETRIES):
                  try:
                      state = {"closing": False}
    
                      async def _stream_input(
                          frames: list[rtc.AudioFrame], stream: RecognizeStream, state: dict = state
                      ):
                          for frame in frames:
                              stream.push_frame(frame)
                              await asyncio.sleep(0.005)
    
                          stream.end_input()
                          state["closing"] = True
    
                      async def _stream_output(stream: RecognizeStream, state: dict = state):
                          text = ""
                          # make sure the events are sent in the right order
                          recv_start, recv_end = False, True
                          start_time = time.time()
                          got_final_transcript = False
                          sos_count, final_count = 0, 0
    
                          async for event in stream:
                              if event.ty
  

• tests.test_stt::test_stream[livekit.agents.inference]

  stt_factory = <function parameter_factory.<locals>.<lambda> at 0x7fdcae324ca0>
  request = <FixtureRequest for <Coroutine test_stream[livekit.agents.inference]>>
  
      @pytest.mark.usefixtures("job_process")
      @pytest.mark.parametrize("stt_factory", STTs + STREAM_ONLY_STTs)
      async def test_stream(stt_factory: Callable[[], STT], request):
          sample_rate = SAMPLE_RATE
          plugin_id = request.node.callspec.id.split("-")[0]
          frames, transcript, _ = await make_test_speech(chunk_duration_ms=10, sample_rate=sample_rate)
    
          # TODO: differentiate missing key vs other errors
          try:
              stt_instance: STT = stt_factory()
          except ValueError as e:
              pytest.skip(f"{plugin_id}: {e}")
    
          async with stt_instance as stt:
              label = f"{stt.model}@{stt.provider}"
              if not stt.capabilities.streaming:
                  pytest.skip(f"{label} does not support streaming")
    
              for attempt in range(MAX_RETRIES):
                  try:
                      state = {"closing": False}
    
                      async def _stream_input(
                          frames: list[rtc.AudioFrame], stream: RecognizeStream, state: dict = state
                      ):
                          for frame in frames:
                              stream.push_frame(frame)
                              await asyncio.sleep(0.005)
    
                          stream.end_input()
                          state["closing"] = True
    
                      async def _stream_output(stream: RecognizeStream, state: dict = state):
                          text = ""
                          # make sure the events are sent in the right order
                          recv_start, recv_end = False, True
                          start_time = time.time()
                          got_final_transcript = False
                          sos_count, final_count = 0, 0
    
                          async for event in stream:
                              if event.
  

Skipped Tests

Test	Reason
tests.test_stt::test_recognize[livekit.plugins.assemblyai]	universal-3-5-pro@AssemblyAI does not support batch recognition
tests.test_stt::test_recognize[livekit.plugins.speechmatics]	enhanced@Speechmatics does not support batch recognition
tests.test_stt::test_recognize[livekit.plugins.fireworksai]	unknown@FireworksAI does not support batch recognition
tests.test_stt::test_recognize[livekit.plugins.nvidia]	unknown@unknown does not support batch recognition
tests.test_stt::test_recognize[livekit.plugins.cartesia]	ink-2@Cartesia does not support batch recognition
tests.test_stt::test_recognize[livekit.plugins.soniox]	stt-rt-v5@Soniox does not support batch recognition
tests.test_stt::test_recognize[livekit.agents.inference]	unknown@livekit does not support batch recognition
tests.test_stt::test_recognize[livekit.plugins.azure]	unknown@Azure STT does not support batch recognition
tests.test_stt::test_recognize[livekit.plugins.cartesia._legacy]	ink-whisper@Cartesia does not support batch recognition
tests.test_stt::test_recognize[livekit.plugins.aws]	unknown@Amazon Transcribe does not support batch recognition
tests.test_stt::test_recognize[livekit.plugins.deepgram.STTv2]	flux-general-en@Deepgram does not support batch recognition
tests.test_stt::test_stream[livekit.plugins.elevenlabs]	scribe_v1@ElevenLabs does not support streaming
tests.test_stt::test_recognize[livekit.plugins.gradium.STT]	unknown@Gradium does not support batch recognition
tests.test_stt::test_stream[livekit.plugins.fal]	Wizper@Fal does not support streaming
tests.test_stt::test_stream[livekit.plugins.mistralai]	voxtral-mini-latest@MistralAI does not support streaming
tests.test_stt::test_stream[livekit.plugins.openai]	gpt-4o-mini-transcribe@api.openai.com does not support streaming

---

Triggered by workflow run #3677

[chenghao-mou]

/test-stt

[devin-ai-integration]

Devin Review found 1 new potential issue.

[devin-ai-integration]

🚩 cibuildwheel build filter missing for blockguard and blingfire

The livekit-durable cibuildwheel config explicitly lists build = "cp310-* cp311-* cp312-* cp313-* cp314-*", but livekit-blockguard and livekit-blingfire have no build key. This means they rely on cibuildwheel's default build matrix. This is pre-existing and not introduced by this PR, but now that the host Python is 3.14, it's worth verifying that the default cibuildwheel matrix includes cp314 wheels if that's desired for these packages.

---

Was this helpful? React with 👍 or 👎 to provide feedback.

[devin-ai-integration]

Devin Review found 1 new potential issue.

[devin-ai-integration]

🚩 Unit and integration tests no longer run on any Python version below 3.14

Prior to this PR, unit-tests, blockguard-tests, evaluation, evals, test-realtime, test-stt, and download-stats all ran on Python 3.12 (or 3.10/3.11 for some build jobs). After this PR, they all run exclusively on 3.14. The only CI job that still exercises an older Python is the type-check matrix in ci.yml, which covers 3.10 and 3.13. This means runtime regressions specific to Python 3.10–3.13 (e.g. behavioral differences in asyncio, typing, stdlib changes) would no longer be caught by tests — only type-checking. This is a deliberate tradeoff but worth noting since AGENTS.md:125 states "Python 3.10+ compatibility required".

---

Was this helpful? React with 👍 or 👎 to provide feedback.

[devin-ai-integration]

Devin Review found 1 new potential issue.

View 6 additional findings in Devin Review.

[devin-ai-integration]

🟡 Published API documentation will silently lose the BitHuman avatar plugin pages

The documentation build is switched to run on Python 3.14 (python-version: "3.14" at .github/workflows/publish-docs.yml:22), a version on which the BitHuman integration cannot be loaded, so its reference pages vanish from the published docs site.
Impact: Users browsing the Python API reference will no longer find any BitHuman avatar documentation after the next docs publish.

Mechanism: dependency marker excludes 3.14, import fails, pdoc --skip-errors drops the module, S3 sync --delete removes the old pages

• livekit-plugins/livekit-plugins-bithuman/pyproject.toml:35 declares bithuman>=0.5.25,<3; python_version >= '3.11' and python_version < '3.14', so uv sync --all-extras --group docs on 3.14 installs the plugin package without its bithuman dependency.
• livekit-plugins/livekit-plugins-bithuman/livekit/plugins/bithuman/__init__.py:22-27 imports .avatar and re-raises ImportError when the dependency is missing.
• The docs step runs pdoc --skip-errors --html ... livekit (.github/workflows/publish-docs.yml:36), which silently skips modules that fail to import, so livekit.plugins.bithuman produces no HTML/markdown.
• The upload step uses aws s3 sync --delete (.github/workflows/publish-docs.yml:44-45), so the previously published bithuman pages get deleted from the docs bucket.
• Previously this job ran on Python 3.12, where the marker held and the module imported fine.

Suggested change

	python-version: "3.14"
	python-version: "3.13"

---

Was this helpful? React with 👍 or 👎 to provide feedback.