initial commit

.idea/.gitignore

@@ -0,0 +1,8 @@
+# Default ignored files
+/shelf/
+/workspace.xml
+# Editor-based HTTP Client requests
+/httpRequests/
+# Datasource local storage ignored files
+/dataSources/
+/dataSources.local.xml

README.md

@@ -0,0 +1,140 @@
+---
+language:
+  - en
+  - zh
+  - de
+  - es
+  - ru
+  - ko
+  - fr
+  - ja
+  - pt
+  - tr
+  - pl
+  - ca
+  - nl
+  - ar
+  - sv
+  - it
+  - id
+  - hi
+  - fi
+  - vi
+  - he
+  - uk
+  - el
+  - ms
+  - cs
+  - ro
+  - da
+  - hu
+  - ta
+  - 'no'
+  - th
+  - ur
+  - hr
+  - bg
+  - lt
+  - la
+  - mi
+  - ml
+  - cy
+  - sk
+  - te
+  - fa
+  - lv
+  - bn
+  - sr
+  - az
+  - sl
+  - kn
+  - et
+  - mk
+  - br
+  - eu
+  - is
+  - hy
+  - ne
+  - mn
+  - bs
+  - kk
+  - sq
+  - sw
+  - gl
+  - mr
+  - pa
+  - si
+  - km
+  - sn
+  - yo
+  - so
+  - af
+  - oc
+  - ka
+  - be
+  - tg
+  - sd
+  - gu
+  - am
+  - yi
+  - lo
+  - uz
+  - fo
+  - ht
+  - ps
+  - tk
+  - nn
+  - mt
+  - sa
+  - lb
+  - my
+  - bo
+  - tl
+  - mg
+  - as
+  - tt
+  - haw
+  - ln
+  - ha
+  - ba
+  - jw
+  - su
+tags:
+  - audio
+  - automatic-speech-recognition
+license: mit
+library_name: ctranslate2
+---
+
+# Whisper large-v2 model for CTranslate2
+
+This repository contains the conversion of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) to the [CTranslate2](https://github.com/OpenNMT/CTranslate2) model format.
+
+This model can be used in CTranslate2 or projects based on CTranslate2 such as [faster-whisper](https://github.com/guillaumekln/faster-whisper).
+
+## Example
+
+```python
+from faster_whisper import WhisperModel
+
+model = WhisperModel("large-v2")
+
+segments, info = model.transcribe("audio.mp3")
+for segment in segments:
+    print("[%.2fs -> %.2fs] %s" % (segment.start, segment.end, segment.text))
+```
+
+## Conversion details
+
+The original model was converted with the following command:
+
+```
+ct2-transformers-converter --model openai/whisper-large-v2 --output_dir faster-whisper-large-v2 \
+    --copy_files tokenizer.json --quantization float16
+```
+
+Note that the model weights are saved in FP16. This type can be changed when the model is loaded using the [`compute_type` option in CTranslate2](https://opennmt.net/CTranslate2/quantization.html).
+
+## More information
+
+**For more information about the original model, see its [model card](https://huggingface.co/openai/whisper-large-v2).**

config.json

@@ -0,0 +1,291 @@
+{
+  "alignment_heads": [
+    [
+      10,
+      12
+    ],
+    [
+      13,
+      17
+    ],
+    [
+      16,
+      11
+    ],
+    [
+      16,
+      12
+    ],
+    [
+      16,
+      13
+    ],
+    [
+      17,
+      15
+    ],
+    [
+      17,
+      16
+    ],
+    [
+      18,
+      4
+    ],
+    [
+      18,
+      11
+    ],
+    [
+      18,
+      19
+    ],
+    [
+      19,
+      11
+    ],
+    [
+      21,
+      2
+    ],
+    [
+      21,
+      3
+    ],
+    [
+      22,
+      3
+    ],
+    [
+      22,
+      9
+    ],
+    [
+      22,
+      12
+    ],
+    [
+      23,
+      5
+    ],
+    [
+      23,
+      7
+    ],
+    [
+      23,
+      13
+    ],
+    [
+      25,
+      5
+    ],
+    [
+      26,
+      1
+    ],
+    [
+      26,
+      12
+    ],
+    [
+      27,
+      15
+    ]
+  ],
+  "lang_ids": [
+    50259,
+    50260,
+    50261,
+    50262,
+    50263,
+    50264,
+    50265,
+    50266,
+    50267,
+    50268,
+    50269,
+    50270,
+    50271,
+    50272,
+    50273,
+    50274,
+    50275,
+    50276,
+    50277,
+    50278,
+    50279,
+    50280,
+    50281,
+    50282,
+    50283,
+    50284,
+    50285,
+    50286,
+    50287,
+    50288,
+    50289,
+    50290,
+    50291,
+    50292,
+    50293,
+    50294,
+    50295,
+    50296,
+    50297,
+    50298,
+    50299,
+    50300,
+    50301,
+    50302,
+    50303,
+    50304,
+    50305,
+    50306,
+    50307,
+    50308,
+    50309,
+    50310,
+    50311,
+    50312,
+    50313,
+    50314,
+    50315,
+    50316,
+    50317,
+    50318,
+    50319,
+    50320,
+    50321,
+    50322,
+    50323,
+    50324,
+    50325,
+    50326,
+    50327,
+    50328,
+    50329,
+    50330,
+    50331,
+    50332,
+    50333,
+    50334,
+    50335,
+    50336,
+    50337,
+    50338,
+    50339,
+    50340,
+    50341,
+    50342,
+    50343,
+    50344,
+    50345,
+    50346,
+    50347,
+    50348,
+    50349,
+    50350,
+    50351,
+    50352,
+    50353,
+    50354,
+    50355,
+    50356,
+    50357
+  ],
+  "suppress_ids": [
+    1,
+    2,
+    7,
+    8,
+    9,
+    10,
+    14,
+    25,
+    26,
+    27,
+    28,
+    29,
+    31,
+    58,
+    59,
+    60,
+    61,
+    62,
+    63,
+    90,
+    91,
+    92,
+    93,
+    359,
+    503,
+    522,
+    542,
+    873,
+    893,
+    902,
+    918,
+    922,
+    931,
+    1350,
+    1853,
+    1982,
+    2460,
+    2627,
+    3246,
+    3253,
+    3268,
+    3536,
+    3846,
+    3961,
+    4183,
+    4667,
+    6585,
+    6647,
+    7273,
+    9061,
+    9383,
+    10428,
+    10929,
+    11938,
+    12033,
+    12331,
+    12562,
+    13793,
+    14157,
+    14635,
+    15265,
+    15618,
+    16553,
+    16604,
+    18362,
+    18956,
+    20075,
+    21675,
+    22520,
+    26130,
+    26161,
+    26435,
+    28279,
+    29464,
+    31650,
+    32302,
+    32470,
+    36865,
+    42863,
+    47425,
+    49870,
+    50254,
+    50258,
+    50358,
+    50359,
+    50360,
+    50361,
+    50362
+  ],
+  "suppress_ids_begin": [
+    220,
+    50257
+  ]
+}

handler.py

@@ -0,0 +1,43 @@
+import io
+import base64
+from faster_whisper import WhisperModel
+import logging
+
+logging.basicConfig(level=logging.DEBUG)
+
+
+class EndpointHandler:
+    def __init__(self, path=""):
+        self.model = WhisperModel("large-v2", num_workers=30)
+
+    def __call__(self, data: dict[str, str]):
+        # process inputs
+        inputs = data.pop("inputs", data)
+        language = data.pop("language", "de")
+        task = data.pop("task", "transcribe")
+
+        # Decode base64 string to bytes
+        audio_bytes_decoded = base64.b64decode(inputs)
+        logging.debug(f"Decoded Bytes Length: {len(audio_bytes_decoded)}")
+        audio_bytes = io.BytesIO(audio_bytes_decoded)
+
+        # run inference pipeline
+        logging.info("Running inference...")
+        segments, info = self.model.transcribe(audio_bytes, language=language, task=task)
+
+        # postprocess the prediction
+        full_text = []
+        for segment in segments:
+            full_text.append({"segmentId": segment.id,
+                              "text": segment.text,
+                              "timestamps": {
+                                  "start": segment.start,
+                                  "end": segment.end
+                              }
+                              })
+
+            if segment.id % 100 == 0:
+                logging.info("segment " + str(segment.id) + " transcribed")
+        logging.info("Inference completed.")
+
+        return full_text