Is it only me or is ElevenLabs TTS throwing an err...
# agentsn
narrow-train-3059
12/09/2025, 4:16 AMIs it only me or is ElevenLabs TTS throwing an error like:
Copy code
Traceback (most recent call last):
File "/Users/xx/Documents/livekitonly/agents/livekit-agents/livekit/agents/tts/tts.py", line 407, in _main_task
raise APIError(f"no audio frames were pushed for text: {self._pushed_text}")
livekit.agents._exceptions.APIError: no audio frames were pushed for text: Hello (body=None, retryable=True)l
lively-winter-88668
12/09/2025, 10:54 AMI thought those kind of errors usually come from some error in the agent code. I don't see any any issues on ElevenLabs and their status page shows green: https://status.elevenlabs.io/
lively-winter-88668
12/09/2025, 11:32 AMPossibly related: https://github.com/livekit/agents/issues/4135
n
narrow-train-3059
12/09/2025, 10:29 PMI am getting that and this error continuously:
Copy code
WARNI… livekit.plugins… websocket closed unexpectedly
WARNI… livekit.agents failed to synthesize speech, retrying in {"tts":
0.1s "livekit.plugins.elevenlabs.tts.TTS",
Traceback (most recent call last): "attempt": 1, "streamed": true}
File
"/Users/dhruvladia/Documents/sarvamagain/p
ython-agents-examples/venv/lib/python3.11/
site-packages/livekit/agents/tts/tts.py",
line 383, in _main_task
await self._run(output_emitter)
File
"/Users/dhruvladia/Documents/sarvamagain/p
ython-agents-examples/venv/lib/python3.11/
site-packages/livekit/plugins/elevenlabs/t
ts.py", line 440, in _run
raise e
File
"/Users/dhruvladia/Documents/sarvamagain/p
ython-agents-examples/venv/lib/python3.11/
site-packages/livekit/plugins/elevenlabs/t
ts.py", line 435, in _run
await waiter
livekit.agents._exceptions.APIStatusError:
connection closed (status_code=-1,
request_id=None, body=None,
retryable=True)l
lively-winter-88668
12/10/2025, 9:48 AMI see what you mean with the
pipeline-tts/elevenlabs_tts.pyencoding="pcm_44100"n
narrow-train-3059
12/10/2025, 10:16 AMAfter removing that line, I was receiving this error:
Copy code
failed to synthesize speech, retrying in {"tts":
2.0s "livekit.plugins.elevenlabs.tts.TTS",
Traceback (most recent call last): "attempt": 3, "streamed": true}
File
"/Users/dhruvladia/Documents/sarvamagain/p
ython-agents-examples/venv/lib/python3.11/
site-packages/livekit/agents/tts/tts.py",
line 394, in _main_task
raise APIError(f"no audio frames were
pushed for text: {self._pushed_text}")
livekit.agents._exceptions.APIError: no
audio frames were pushed for text: Hello!
How can I assist you today? (body=None,
retryable=True)l
lively-winter-88668
12/10/2025, 10:26 AMThat's the same error as before, no? So, removing the line had no effect for you? It's fairly consistent for me that removing that line solves the issue. I do notice we're on a different version of livekit-agents, I'm on 1.3.6.
Alternatively, inference works: https://docs.livekit.io/agents/models/tts/inference/elevenlabs/
n
narrow-train-3059
12/10/2025, 11:17 AMI updated to 1.3.6 but still seeing this error:
Copy code
(venv) dhruvladia@Dhruvs-MacBook-Pro python-agents-examples % python pipeline-tts/elevenlabs_tts.py console
Agents Starting console mode 🚀
16:38:47.025 DEBUG asyncio Using selector: KqueueSelector
16:38:47.026 INFO livekit.agents starting worker {"version": "1.3.6", "rtc-version": "1.0.19"}
16:38:47.028 INFO livekit.agents HTTP server listening on :56308
16:38:47.035 INFO livekit.agents initializing job runner {"tid": 33744691}
16:38:47.036 DEBUG asyncio Using selector: KqueueSelector
16:38:47.037 INFO livekit.agents job runner initialized {"tid": 33744691, "elapsed_time": 0.0}
16:38:47.181 DEBUG livekit.agents http_session(): creating a new httpclient ctx
16:38:47.183 DEBUG livekit.agents using audio io: `Console` -> `AgentSession` -> `Console`
WARNI… livekit.agents resume_false_interruption is enabled but audio output does not support pause,
it will be ignored
{"audio_output": "Console"}
16:38:47.184 DEBUG livekit.agents using transcript io: `AgentSession` -> (none)
16:38:47.806 WARNI… livekit.agents failed to synthesize speech, retrying in 0.1s
{"tts": "livekit.plugins.elevenlabs.tts.TTS", "attempt": 1, "streamed": true}
Traceback (most recent call last):
File
"/Users/dhruvladia/Documents/sarvamagain/python-agents-examples/venv/lib/python3.11/site-packages/livekit/agents/tts/tts.py"
, line 407, in _main_task
raise APIError(f"no audio frames were pushed for text: {self._pushed_text}")
livekit.agents._exceptions.APIError: no audio frames were pushed for text: Hi there! Is there anything I can help you with?
(body=None, retryable=True)l
lively-winter-88668
12/10/2025, 11:35 AMOr, from https://docs.livekit.io/agents/models/tts/inference/elevenlabs/, you can replace the TTS call with inference:
tts="elevenlabs/eleven_turbo_v2_5:Xb7hH8MSUJpSbSDYk0k2",n
narrow-train-3059
12/10/2025, 11:50 AMInferencing is working but why not without it?
l
lively-winter-88668
12/10/2025, 11:57 AMI'm not sure, my working theory is that
encoding="pcm_44100"mp3_44100_32m
millions-dog-30135
12/16/2025, 5:22 PMthe error is because the credit limit of your elevenlab api key is completed change the elevenlab api key and check once also make sure you give correct naming ELEVEN_API_KEY="********" like this
r
ripe-account-5250
12/29/2025, 7:45 PMit looks like streaming TTS retries are broken. The retry loop reuses the same SynthesizeStream instance, but the _input_ch channel and plugin state are one-shot resources that get consumed during the first attempt.
As a result, retrying the same request sends no text or produces no segments, leading to consistent failures like APIError: no audio frames were pushed or segments mismatch.
I think this PR should fix that : https://github.com/livekit/agents/pull/4410