Updating wav2vec2_inference for batched_input

[tchawada]

Updating wav2vec2 for batched_input and adding args for parsing batch_size, ctx_len and num_devices

model.compile(batch_size=args.batch_size,
              num_devices=args.num_devices,
              seq_len=args.seq_len,
              num_cores=args.num_cores)

[quic-hemagnih]

@tchawada Can you please fix the LINT issues.

[quic-hemagnih]

LGTM, lets merge once CI passes