From 3beee3cfe97f3b3c04458ba5ff2612c8bd9358a7 Mon Sep 17 00:00:00 2001
From: moe-berri <moe@berri.ai>
Date: Mon, 26 Oct 2026 13:42:01 -0700
Subject: [PATCH] fix(elevenlabs): map wav and pcm TTS formats to 24kHz output
 formats (#48455)

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---
 .../text_to_speech/transformation.py          |  4 +--
 .../test_transformation.py                    |  2 +-
 ...levenlabs_text_to_speech_transformation.py | 28 +++++++++++++++++++
 3 files changed, 31 insertions(+), 3 deletions(-)

diff --git a/litellm/llms/elevenlabs/text_to_speech/transformation.py b/litellm/llms/elevenlabs/text_to_speech/transformation.py
index f910ef8c057..23ccf41c9da 100644
--- a/litellm/llms/elevenlabs/text_to_speech/transformation.py
+++ b/litellm/llms/elevenlabs/text_to_speech/transformation.py
@@ -56,9 +56,9 @@ class ElevenLabsTextToSpeechConfig(BaseTextToSpeechConfig):
     # Response format mappings from OpenAI to ElevenLabs
     FORMAT_MAPPINGS = {
         "mp3": "mp3_44100_128",
-        "pcm": "pcm_44100",
+        "pcm": "pcm_24000",
         "opus": "opus_48000_128",
-        # ElevenLabs does not support WAV, AAC, or FLAC formats.
+        "wav": "wav_24000",
     }
 
     ELEVENLABS_QUERY_PARAMS_KEY = "__elevenlabs_query_params__"
diff --git a/tests/unit/llms/elevenlabs/audio_transcription/test_transformation.py b/tests/unit/llms/elevenlabs/audio_transcription/test_transformation.py
index 86252a1a95d..f55644eccae 100644
--- a/tests/unit/llms/elevenlabs/audio_transcription/test_transformation.py
+++ b/tests/unit/llms/elevenlabs/audio_transcription/test_transformation.py
@@ -314,4 +314,4 @@ class TestElevenLabsTextToSpeechTransformation:
         )
 
         assert voice_id in url
-        assert "output_format=pcm_44100" in url
+        assert "output_format=pcm_24000" in url
diff --git a/tests/unit/llms/elevenlabs/test_elevenlabs_text_to_speech_transformation.py b/tests/unit/llms/elevenlabs/test_elevenlabs_text_to_speech_transformation.py
index d05371d7df9..88f0ae1d524 100644
--- a/tests/unit/llms/elevenlabs/test_elevenlabs_text_to_speech_transformation.py
+++ b/tests/unit/llms/elevenlabs/test_elevenlabs_text_to_speech_transformation.py
@@ -55,3 +55,31 @@ def test_speech_keeps_an_internal_prefixed_kwarg_out_of_the_elevenlabs_request(r
     sent: Final = json.loads(respx_mock.calls[0].request.content)
     assert "_litellm_undeclared_sentinel" not in sent, sent
     assert sent["text"] == "hi"
+
+
+@pytest.mark.parametrize(
+    ("response_format", "expected_output_format"),
+    [("wav", "wav_24000"), ("pcm", "pcm_24000")],
+)
+def test_speech_maps_wav_and_pcm_to_elevenlabs_output_formats(
+    respx_mock: respx.MockRouter,
+    response_format: str,
+    expected_output_format: str,
+) -> None:
+    api_base: Final = "http://localhost:12346"
+    mock_route: Final = respx_mock.post(url__regex=rf"{api_base}/v1/text-to-speech/.*").mock(
+        return_value=httpx.Response(status_code=200, content=b"audio", headers={"content-type": "audio/mpeg"})
+    )
+
+    litellm.speech(
+        model="elevenlabs/eleven_multilingual_v2",
+        input="hi",
+        voice="21m00Tcm4TlvDq8ikWAM",
+        response_format=response_format,
+        api_base=api_base,
+        api_key="fake_elevenlabs_api_key",
+    )
+
+    assert mock_route.called
+    request_url: Final = httpx.URL(str(respx_mock.calls[0].request.url))
+    assert request_url.params["output_format"] == expected_output_format