Afrinetwork7
commited on
Commit
•
488f2ba
1
Parent(s):
521a4ba
Update asr.py
Browse files
asr.py
CHANGED
@@ -3,76 +3,19 @@ from transformers import Wav2Vec2ForCTC, AutoProcessor
|
|
3 |
import torch
|
4 |
import numpy as np
|
5 |
from pathlib import Path
|
6 |
-
|
7 |
-
from huggingface_hub import hf_hub_download
|
8 |
-
from torchaudio.models.decoder import ctc_decoder
|
9 |
|
10 |
ASR_SAMPLING_RATE = 16_000
|
11 |
-
|
12 |
-
|
13 |
-
with open(f"data/asr/all_langs.tsv") as f:
|
14 |
-
for line in f:
|
15 |
-
iso, name = line.split(" ", 1)
|
16 |
-
ASR_LANGUAGES[iso.strip()] = name.strip()
|
17 |
|
18 |
MODEL_ID = "facebook/mms-1b-all"
|
19 |
|
20 |
processor = AutoProcessor.from_pretrained(MODEL_ID)
|
21 |
model = Wav2Vec2ForCTC.from_pretrained(MODEL_ID)
|
22 |
|
23 |
-
|
24 |
-
# lm_decoding_config = {}
|
25 |
-
# lm_decoding_configfile = hf_hub_download(
|
26 |
-
# repo_id="facebook/mms-cclms",
|
27 |
-
# filename="decoding_config.json",
|
28 |
-
# subfolder="mms-1b-all",
|
29 |
-
# )
|
30 |
-
|
31 |
-
# with open(lm_decoding_configfile) as f:
|
32 |
-
# lm_decoding_config = json.loads(f.read())
|
33 |
-
|
34 |
-
# # allow language model decoding for "eng"
|
35 |
-
|
36 |
-
# decoding_config = lm_decoding_config["eng"]
|
37 |
-
|
38 |
-
# lm_file = hf_hub_download(
|
39 |
-
# repo_id="facebook/mms-cclms",
|
40 |
-
# filename=decoding_config["lmfile"].rsplit("/", 1)[1],
|
41 |
-
# subfolder=decoding_config["lmfile"].rsplit("/", 1)[0],
|
42 |
-
# )
|
43 |
-
# token_file = hf_hub_download(
|
44 |
-
# repo_id="facebook/mms-cclms",
|
45 |
-
# filename=decoding_config["tokensfile"].rsplit("/", 1)[1],
|
46 |
-
# subfolder=decoding_config["tokensfile"].rsplit("/", 1)[0],
|
47 |
-
# )
|
48 |
-
# lexicon_file = None
|
49 |
-
# if decoding_config["lexiconfile"] is not None:
|
50 |
-
# lexicon_file = hf_hub_download(
|
51 |
-
# repo_id="facebook/mms-cclms",
|
52 |
-
# filename=decoding_config["lexiconfile"].rsplit("/", 1)[1],
|
53 |
-
# subfolder=decoding_config["lexiconfile"].rsplit("/", 1)[0],
|
54 |
-
# )
|
55 |
-
|
56 |
-
# beam_search_decoder = ctc_decoder(
|
57 |
-
# lexicon=lexicon_file,
|
58 |
-
# tokens=token_file,
|
59 |
-
# lm=lm_file,
|
60 |
-
# nbest=1,
|
61 |
-
# beam_size=500,
|
62 |
-
# beam_size_token=50,
|
63 |
-
# lm_weight=float(decoding_config["lmweight"]),
|
64 |
-
# word_score=float(decoding_config["wordscore"]),
|
65 |
-
# sil_score=float(decoding_config["silweight"]),
|
66 |
-
# blank_token="<s>",
|
67 |
-
# )
|
68 |
-
|
69 |
-
|
70 |
-
def transcribe(audio_data=None, lang="eng (English)"):
|
71 |
-
if audio_data is None or (isinstance(audio_data, np.ndarray) and audio_data.size == 0):
|
72 |
-
return "<<ERROR: Empty Audio Input>>"
|
73 |
-
|
74 |
if isinstance(audio_data, tuple):
|
75 |
-
# microphone
|
76 |
sr, audio_samples = audio_data
|
77 |
audio_samples = (audio_samples / 32768.0).astype(np.float32)
|
78 |
if sr != ASR_SAMPLING_RATE:
|
@@ -80,59 +23,57 @@ def transcribe(audio_data=None, lang="eng (English)"):
|
|
80 |
audio_samples, orig_sr=sr, target_sr=ASR_SAMPLING_RATE
|
81 |
)
|
82 |
elif isinstance(audio_data, np.ndarray):
|
83 |
-
# Assuming audio_data is already in the correct format
|
84 |
audio_samples = audio_data
|
85 |
elif isinstance(audio_data, str):
|
86 |
-
# file upload
|
87 |
audio_samples = librosa.load(audio_data, sr=ASR_SAMPLING_RATE, mono=True)[0]
|
88 |
else:
|
89 |
-
|
90 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
91 |
|
92 |
lang_code = lang.split()[0]
|
93 |
processor.tokenizer.set_target_lang(lang_code)
|
94 |
model.load_adapter(lang_code)
|
95 |
|
96 |
-
|
97 |
-
audio_samples, sampling_rate=ASR_SAMPLING_RATE, return_tensors="pt"
|
98 |
-
)
|
99 |
-
|
100 |
-
# set device
|
101 |
-
if torch.cuda.is_available():
|
102 |
-
device = torch.device("cuda")
|
103 |
-
elif (
|
104 |
-
hasattr(torch.backends, "mps")
|
105 |
-
and torch.backends.mps.is_available()
|
106 |
-
and torch.backends.mps.is_built()
|
107 |
-
):
|
108 |
-
device = torch.device("mps")
|
109 |
-
else:
|
110 |
-
device = torch.device("cpu")
|
111 |
-
|
112 |
model.to(device)
|
113 |
-
inputs = inputs.to(device)
|
114 |
|
115 |
-
|
116 |
-
|
117 |
|
118 |
-
|
119 |
-
|
120 |
-
|
121 |
-
|
122 |
-
|
123 |
-
|
124 |
-
# transcription = " ".join(beam_search_result[0][0].words).strip()
|
125 |
-
|
126 |
-
return transcription
|
127 |
|
|
|
128 |
|
|
|
129 |
ASR_EXAMPLES = [
|
130 |
["upload/english.mp3", "eng (English)"],
|
131 |
# ["upload/tamil.mp3", "tam (Tamil)"],
|
132 |
# ["upload/burmese.mp3", "mya (Burmese)"],
|
133 |
]
|
134 |
|
135 |
-
|
136 |
-
|
137 |
-
|
138 |
-
|
|
|
|
3 |
import torch
|
4 |
import numpy as np
|
5 |
from pathlib import Path
|
6 |
+
import concurrent.futures
|
|
|
|
|
7 |
|
8 |
ASR_SAMPLING_RATE = 16_000
|
9 |
+
CHUNK_LENGTH_S = 60 # Increased to 60 seconds per chunk
|
10 |
+
MAX_CONCURRENT_CHUNKS = 4 # Adjust based on VRAM availability
|
|
|
|
|
|
|
|
|
11 |
|
12 |
MODEL_ID = "facebook/mms-1b-all"
|
13 |
|
14 |
processor = AutoProcessor.from_pretrained(MODEL_ID)
|
15 |
model = Wav2Vec2ForCTC.from_pretrained(MODEL_ID)
|
16 |
|
17 |
+
def load_audio(audio_data):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
if isinstance(audio_data, tuple):
|
|
|
19 |
sr, audio_samples = audio_data
|
20 |
audio_samples = (audio_samples / 32768.0).astype(np.float32)
|
21 |
if sr != ASR_SAMPLING_RATE:
|
|
|
23 |
audio_samples, orig_sr=sr, target_sr=ASR_SAMPLING_RATE
|
24 |
)
|
25 |
elif isinstance(audio_data, np.ndarray):
|
|
|
26 |
audio_samples = audio_data
|
27 |
elif isinstance(audio_data, str):
|
|
|
28 |
audio_samples = librosa.load(audio_data, sr=ASR_SAMPLING_RATE, mono=True)[0]
|
29 |
else:
|
30 |
+
raise ValueError(f"Invalid Audio Input Instance: {type(audio_data)}")
|
31 |
+
return audio_samples
|
32 |
+
|
33 |
+
def process_chunk(chunk, device):
|
34 |
+
inputs = processor(chunk, sampling_rate=ASR_SAMPLING_RATE, return_tensors="pt").to(device)
|
35 |
+
with torch.no_grad():
|
36 |
+
outputs = model(**inputs).logits
|
37 |
+
ids = torch.argmax(outputs, dim=-1)[0]
|
38 |
+
return processor.decode(ids)
|
39 |
+
|
40 |
+
def transcribe(audio_data=None, lang="eng (English)"):
|
41 |
+
if audio_data is None or (isinstance(audio_data, np.ndarray) and audio_data.size == 0):
|
42 |
+
return "<<ERROR: Empty Audio Input>>"
|
43 |
+
|
44 |
+
try:
|
45 |
+
audio_samples = load_audio(audio_data)
|
46 |
+
except Exception as e:
|
47 |
+
return f"<<ERROR: {str(e)}>>"
|
48 |
|
49 |
lang_code = lang.split()[0]
|
50 |
processor.tokenizer.set_target_lang(lang_code)
|
51 |
model.load_adapter(lang_code)
|
52 |
|
53 |
+
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
54 |
model.to(device)
|
|
|
55 |
|
56 |
+
chunk_length = int(CHUNK_LENGTH_S * ASR_SAMPLING_RATE)
|
57 |
+
chunks = [audio_samples[i:i+chunk_length] for i in range(0, len(audio_samples), chunk_length)]
|
58 |
|
59 |
+
transcriptions = []
|
60 |
+
|
61 |
+
with concurrent.futures.ThreadPoolExecutor(max_workers=MAX_CONCURRENT_CHUNKS) as executor:
|
62 |
+
future_to_chunk = {executor.submit(process_chunk, chunk, device): chunk for chunk in chunks}
|
63 |
+
for future in concurrent.futures.as_completed(future_to_chunk):
|
64 |
+
transcriptions.append(future.result())
|
|
|
|
|
|
|
65 |
|
66 |
+
return " ".join(transcriptions)
|
67 |
|
68 |
+
# Example usage
|
69 |
ASR_EXAMPLES = [
|
70 |
["upload/english.mp3", "eng (English)"],
|
71 |
# ["upload/tamil.mp3", "tam (Tamil)"],
|
72 |
# ["upload/burmese.mp3", "mya (Burmese)"],
|
73 |
]
|
74 |
|
75 |
+
if __name__ == "__main__":
|
76 |
+
for audio_file, language in ASR_EXAMPLES:
|
77 |
+
print(f"Transcribing {audio_file} in {language}")
|
78 |
+
transcription = transcribe(audio_file, language)
|
79 |
+
print(f"Transcription: {transcription}\n")
|